MRI–radiomics–clinical–based nomogram for prediction postpartum hemorrhage in patients with suspected placenta accrete spectrum before cesarean section

MRI–radiomics–clinical–based nomogram for prediction postpartum hemorrhage in patients with suspected placenta accrete spectrum before cesarean section | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article MRI–radiomics–clinical–based nomogram for prediction postpartum hemorrhage in patients with suspected placenta accrete spectrum before cesarean section Yumin Hu, Yechao Huang, Bo Chen, Di Shen, Xia Li, Zufei Wang, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4550980/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose To develop and validate the nomogram by combining MRI-derived radiomics and clinical features for predicting postpartum hemorrhage in high-risk placenta accreta spectrum (PAS) patients before cesarean section. Methods The T2WI sagittal MR images and clinical data of 70 postpartum hemorrhage (+) pregnant women and 104 postpartum hemorrhage (-) pregnant women were retrospectively collected from two centers. These pregnancies were divided into a training (n = 105), an independent validation (n = 28), and an external validation (n = 41) cohort. Radiomic features were extracted, and radiomics signature were constructed. Clinical features were analyzed retrospectively. The clinical model, the radiomic model, and the clinicoradiomic model were compared. The nomogram of the optimal model was constructed to predict the risk of postpartum hemorrhage.The diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC) and the DeLong test. Decision curve analysis (DCA) was performed to determine the performance of the best predictive model. Results Eighteen radiomic features showed a strong correlation with PPH. Four MRI features were selected as clinical features.The clinicoradiomic model resulted in the best discrimination ability for risk prediction of PPH, with an AUC of 0.956 (95% CI, 0.9101.000), 0.781(95% CI, 0.606 0.955), and 0.702 (95% CI, 0.541 0.864) in the training, independent validation and external validation cohorts respectively. The clinicoradiomic nomogram, incorporating radiomics signature and four MRI features, was developed. The calibration was good and DCA confirmed the clinical utility of the nomogram. Conclusion Obstetricians can use the nomogram to noninvasively predict PPH and guide them in creating reasonable preoperative treatment plans. Magnetic resonance imaging Placenta accreta spectrum Nomogram Postpartum hemorrhage Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Postpartum hemorrhage (PPH) is the leading direct cause of maternal mortality worldwide[ 1 ].PPH is defined as 500 mL or more of blood loss from the genital tract following a normal vaginal delivery (NVD) or 1000 mL or more following a cesarean section within 24 hours of birth[ 2 – 6 ].Postpartum hemorrhage is the most common complication of Placenta Accrete Spectrum (PAS). It may lead to uncontrollable bleeding and threaten the lives of mother and baby[ 7 ]. Early prediction of PPH in suspected PAS patients would aid delivery planning, facilitate the preparation of blood products and transfusion requirements, as well as decrease maternal complications[ 8 ]. At present, the prediction of postpartum hemorrhage in patients with suspected PAS includes laboratory indicators of coagulation function, clinical risk factors of PAS, and evaluation of ultrasound and magnetic resonance. However, these indicators are relatively scattered, and the prediction ability of postpartum hemorrhage is limited. Therefore, we need to build a model with predictive probability to predict postpartum hemorrhage. The development of a predictive model could optimize maternal outcomes by improving clinical recognition and timely treatment to avoid excessive blood loss in women with suspected PAS pregnancies [ 9 ]. Radiomics analysis quantifying high-dimensional tissue features that cannot be observed by the naked eye has shown great potential in precision medicine, aiming to extract mineable high-dimensional data from clinical images, which was followed by subsequent data analysis for decision support[ 10 ]. Although radiomic analysis is mostly applied to tumors, some studies found it to be a feasible tool for PAS diagnosis and even PAS severity assessment[ 8 , 11 – 13 ].However, there are not many studies on prenatal prediction of postpartum hemorrhage in pregnant women suspected of having PAS, which combined radiomic features with clinical, laboratory and MR morphological indicators in their studies[ 8 , 14 , 15 ]. In this retrospective study, we aimed to extract the radiomic features of the placenta that correlated with PPH and at the same time, develop a nomogram for a predictive model that could optimize the maternal outcomes of patients with suspected PAS. Materials and methods 2.1 Patients This retrospective study was approved by the Institutional Review Board of the Lishui Central Hospital (center 1) and the Second Affiliated Hospital and Yuying Children’s Hospital, Wenzhou Medical University (center 2), with written informed consent waived. Pregnant women suspicious of PAS clinically and on ultrasonic examination were collected from the electronic medical record system of Center 1 (n = 155) from September 2016 to April 2022 and Center 2 (n = 60) from July 2020 to January 2022. The inclusion criteria were (a) singleton pregnancies with intraoperative bleeding and bleeding volume recorded in postoperative 24-hour nursing records; (b) pregnancies that underwent MRI before cesarean operation; and (c) pregnancies that had complete pathologic and clinical information. The exclusion criteria were (a) patients with vaginal delivery (n = 3 in center 1; n = 4 in center 2) (b) incomplete clinical data (n = 7 in center 1; n = 8 in center 2) (c) poor image quality(n = 8 in center 1; n = 5 in center 2) (d) fatal congenital dysplasia (n = 3 in center 1; n = 2 in center 2). Finally, there were 133 pregnant women (n = 50 in the PPH(+) group; n = 83 in the PPH(-) group) in center 1 and 41 pregnant women (n = 20 in the PPH(+) group; n = 21 in the PPH(-) group) in center 2 identified in our study. Pregnancies from Center 1 were randomized into a training (n = 105) and an independent validation (n = 28) cohort with a proportion of 8:2. Pregnancies from Center 2 were scheduled as the external validation cohort (n = 41) (Fig. 1). 2.2. MRI examination The pelvic MR image acquisition of all pregnant women was performed on a 1.5-T MRI scanner (MAGENTOM Area; Siemens) in center 1 and a 1.5-T MRI scanner (MAGENTOM Avanto, Siemens) in center 2, both with a 6-channel abdominal phased-array surface coil. The sequence we used for this study is the sagittal FS-T2WI images. The parameters of MRI sequences in these two centers are presented in Supplemental Table 1. 2.3 Radiologic evaluation Two radiologists (CL and ZW with 16 and 21 years of experience in imaging diagnosis) evaluated the image findings separately on a RAD info RSVS system. Both readers were blinded to previous MRI interpretation, previous ultrasound reports, and the surgical and pathologic reports from delivery. Any disagreements were resolved by consensus. Seven MR features, according to the consensus statement of the Society of Abdominal Radiology (SAR) and the European Society of Urogenital Radiology (ESUR),were evaluated and recorded: intraplacental dark T2 bands, placenta bulge, loss of retroplacental T2-hypointense line, myometrial thinning/disruption, bladder tenting sign, increased placental vascularity, and abnormal vascularization of the placental bed[ 16 ]. 2.4 MR image segmentation and radiomic features extraction The DICOM of sagittal T2WI images, which is the optimal position for the uterus and placenta observation, was retrieved from PACS and uploaded to 3D slicer( https://www.slicer.org ). All volume of interest (VOI) segmentation was conducted by a radiologist (CL) who had 16 years of imaging diagnosis experience and was re-checked by SX (15 years of experience); both were blind to the pathology reports. VOI segmentations were circled on the sagittal T2WI of the placenta. Quantitative radiomics features were extracted from VOIs on Pyradiomics(3.0.1). The extracted features were divided into three categories: shape-based features, first-order statistics,and texture features. 2.5 Radiomic and clinical features selection 987 radiomic features were extracted from the placenta volume of interest (VOI), including i) 14 shape features, iii) 203first order features, iiii) 770 texture features. To select the radiomic features highly correlated with PPH, we used least absolute shrinkage and selection operator (LASSO) regression with three-fold cross-validation. For the LASSO algorithm, L1 regularizer was used as the cost function, the error value of cross-validation was 5, and the maximum number of iterations was 1000. The clinical features include the basic data of prenatal patients, laboratory indicators and MR signs. The basic data of the patients included maternal age, gestational age at CD, previous CD history, number of previous abortions, number of abortions, presence of placenta previa, and times of prenatal bleeding. Preoperative laboratory indicators included hemoglobin level, platelet level, prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), D-Dimer, fibrinogen. MR signs included placenta bulge,loss of retroplacental T2-hypointense line,myometrial thinning/disruption,bladder tenting sign,increased placental vascularity,abnormal vascularization of the placental bed. So the LASSO method was also used to obtain the subset of predictors. 2.6 Radiomic and clinicoradiomic model building After feature selection, we built the radiomic model using the support vector machine (SVM) algorithm with a radial basis kernel for risk prediction of PPH based on the selected radiomic features. The area under the receiver operator characteristic (ROC) curve (AUC) was used to validate the performance of the developed radiomics models in training, internal validation and external validation cohorts. The best predictive model was determined, and the probability output by the model was recorded as a radiomics score (Radscore).After that, the selected clinical predictor was added to the radiomic model to form a clinicoradiomic model[ 17 ]. 2.7 Development and validation of the nomogram After model comparison, a nomogram for clinical use was built based on the model with the best discrimination ability for PPH. The calibration curves were used to evaluate the agreement between the observed results and predicted probability. The clinical utility of the nomogram was evaluated by quantifying the net benefit under different threshold probabilities by decision curve analysis. The work flow was shown in Fig. 2. 2.8 Statistical analysis Statistical analysis was performed using SPSS (SPSS 27.0; IBM Corp), MedCalc and R Programming Language (version 3.5.5; http://www.Rproject.org ). The SPSS software was used to perform the variable comparison between different groups or cohorts. Continuous variables were compared by the Student t test or Mann–Whitney U test between PPH(་) and PPH(-) groups. Categorical variables were compared using the 2×2 or R×C χ2 test or Fisher exact test. Predictive performance(AUC, Sensitivity, Specificity, Accuracy, PPV, NPV, P value) of MRI morphological features for PPH were calculated by MedCalc. The R Programming Language was applied to conduct the LASSO regression analysis, ROC analysis (with the area under the curve [AUC] value calculated), calibration curve analysis (with the Hosmer–Lemeshow test), and decision curve analysis. The AUC values among different models were compared using the DeLong method. A p value less than 0.05 was set as the statistical significance threshold. Nomogram development and calibration plots were performed using the “rms” package. The decision curve analysis was performed using the “dca.R.” package. Results 3.1 Clinical characteristics and MR features of patients The clinical characteristics before and after the cesarean section of all pregnant women were summarized in Table 1 . There were statistical differences in gestational age at CD (P = 0.037), history of cesarean section (P = 0.020), number of cesarean section (P = 0.010), depth of PAS invasiveness (P<0.001), whether postpartum blood transfusion performed (P<0.001), preoperative platelet count (P = 0.033), postoperative hemoglobin levels (P = 0.001) and platelet count (P<0.001) between the groups with and without PPH in the training group. There were statistical differences in whether postpartum blood transfusion was performed between the groups with and without PPH in independent validation cohort (P = 0.002) and the external validation cohort (P = 0.000). The diagnostic performance of MRI morphological findings for PAS in the training, independent validation, and external validation cohorts were shown in Table 2 , respectively. Among seven MRI morphological findings, five MRI morphological findings were significantly different between PPH(+) and PPH(-) pregnancies in the training cohort(Intraplacental dark T2 bands P = 0.0001,Placenta bulge P = 0.0001,Loss of retroplacental T2-hypointense line P<0.0001,Myometrial thinning/disruption 0.0352,Increased placental vascularity P = 0.000),one MRI morphological finding significantly different between PPH(+) and PPH(-) pregnancies in the independent validation cohort(Intraplacental dark T2 bands P = 0.0135),and three MRI morphological findings significantly different between PPH(+) and PPH(-) pregnancies in the external validation cohort(Intraplacental dark T2 bands P < 0.0001,Myometrial thinning/disruption P = 0.0022,Increased placental vascularity < 0.0001). 3.2 The clinical and selected radiomic features Four clinical features were selected by LASSO from 20 clinical features(Figure3a,b), all of which are imaging features, including intraplacental dark T2 bands, myometrial thinning/disruption, increased placental vascularity and loss of retroplacental T2-hypointense line. In total, 987 radiomic features were extracted from each VOI of the placenta on T2WI. Among them, 18 features were selected using the LASSO regression algorithm.༈Figure3c-e༉.The 18 radiomic features can be seen in Supplemental table 2. 3.3 Models construction and comparison A radiomic model was constructed using the SVM. The AUC of radiomic signature from the training cohort was 0.929 (95% CI, 0.870, 0.987), from the independent validation cohort were 0.735 (95% CI,, 0.547, 0.922) and from the external validation cohort were 0.686(95% CI, 0.522, 0.850). The SVM model was used to calculate the radiomics signature Rad-score. The Rad-score for each patient presented as a waterfall plot demonstrated significant differences between PPH (+) and PPH (-) in training, internal validation and external validation cohorts (Fig. 4a,c,e). PPH(+) and PPH(-) showed significantly different rad-scores in training, internal validation, and external validation cohorts (Fig. 4b,d,f). The clinicoradiomic model, developed by incorporating radiomic signature with clinical factors, resulted in an AUC of 0.956 (95% CI, 0.911) in the training cohort, 0.781 (95% CI, 0.606 0.955) in the independent validation cohort, 0.702(95% CI, 0.541 0.864) in the external validation cohort respectively for PPH prediction. The performance of radiomic signature, clinical and clinicoradiomic models were shown in Fig. 5 and Table 3 . The performance of the clinicoradiomic models were better than that of the clinical models and radiomic models in all three cohorts. 3.4 Construction and validation of clinicoradiomic nomogram The nomogram was defined based on the clinicoradiomic model Fig. 6a, which showed the best discrimination ability among 3 models. The calibration curve, together with the H-L test, was used to estimate the consistency between the probability of PPH predicted by the clinicoradiomic model and actual outcomes.As shown in Fig. 6b–d, the predicted probability of PPH in the training, independent validation and external validation cohorts respectively match the anticipated probability of PPH. The decision curve evaluated the performance of the clinicoradiomic model in terms of clinical application, hence, reflecting its clinical usefulness. DCA indicated that the farther the decision curve was from the two extreme curves, the higher the clinical decision net benefit. In this study, the net benefit of the clinicoradiomic model were higher than that of the clinical model(training cohort Z = 4.0234,P = 5.73×10 − 5 , independent validation cohort Z = 1.4005,P = 0.1614, external validation cohort Z = 0.093867,P = 0.9252) and the radiomic model(training cohort Z=-1.3121,P = 0.1895, independent validation cohort Z=-0.79029,P = 0.4294, external validation cohort Z=-0.13376,P = 0.8936) (Fig. 6e-g), indicating good discrimination in training, independent validation cohort and external validation cohort. Discussion In this study, radiomic features based on sagittal FS-T2WI placental MR images were selected and used to construct a radiomic model. In order to facilitate visualization, clinical application, and decision-making, we further developed and validated a nomogram that combines clinical and radiomics features for antenatal prediction of postpartum hemorrhage in patients with suspected placenta accrete spectrum. The results showed that the nomogram was superior to the clinical and radiomics models in preoperatively predicting PPH. The precise prenatal diagnosis and prediction of postpartum complications of PAS are increasingly receiving attention from obstetricians[ 18 – 20 ].In recent years, some studies have applied artificial intelligence to the study of PAS[ 21 – 23 ] and its postpartum complications[ 8 , 24 ].There are several differences from previous studies. Firstly, our study had an external validation, which ensured the effectiveness of the training model.Secondly, the patients included in our study included all patients suspected to have PAS by ultrasound, including those confirmed by postoperative pathology, those diagnosed by obstetricians during surgery, and those without PAS. Because we cannot accurately diagnose PAS in pregnant women before surgery, our grouping and constructed model are more in line with clinical needs. In our study, we extracted 18 radiomic features from the training group, and the SVM model based on the 18 radiomic features had good performance in the training cohort(auc = 0.929), independent validation cohort༈auc = 0.735༉and external validation cohort༈auc = 0.686༉. The SVM model we constructed was consistent with previous studies which used in another article on prediction of postpartum haemorrhage. This further confirmed that the SVM model had a good predictive effect on postpartum hemorrhage in the placenta accreta spectrum[ 8 ]. Unlike previous studies, clinical factors related to postpartum hemorrhage, such as prenatal coagulation function, pregnancy history, and prenatal placental MR features, were used together to screen for the best clinical features. And we found that in both baseline table calculations and lasso feature screening, prenatal coagulation function indicators were not statistically significant predictors of postpartum hemorrhage. This may be due to the fact that the prenatal coagulation function of PAS is mostly normal, and abnormalities in coagulation function often occur after PPH during cesarean section[ 25 , 26 ].Finally, four MR features (intraplacental dark T2 bands, increased placental vascularity, myometrial thinning/disruption, loss of retroplacental T2-hypointense line)we screened from 20 clinical features by lasso regression were associated with PPH. And these four MR features were important features associated with postpartum complications in previous studies. Previous studies suggested that intraplacental dark T2 bands bands may help predict placental invasion depth and postpartum hemorrhage in placenta accrete spectrum disorders in high-risk gravid patients[ 7 , 27 , 28 ]. In addition, Bourgioti’s study suggested that increased placental vascularities were valuable for the prediction of postpartum hemorrhage[ 29 ]. Myometrial thinning/disruption and loss of retroplacental T2-hypointense line were also important signs reflecting the severity of PAS[ 30 ].The AUC of clinical model with the combination of these four MR features was 0.734 in training cohort, 0.584 in independent validation cohort,0.694 in external validation cohort. After combining radiomic models with clinical models, the effectiveness of the combined models was significantly improved. The AUC of combined model in training cohort was 0.956, 0.781 in independent validation cohort, 0.702 in external validation cohort. Although it is hard to interpret the specific underlying meaning of radiomic features, radiomic analysis combined with clinical analysis is a useful and stable tool for PPH prediction. The stable performance in different centers also highlights that radiomics analysis can overcome data differences between multiple center. Although our results are promising, there are several limitations to our study. First of all, the number of patients was relatively small. We will increase the sample size and the number of external verifications from other institutions. Secondly, the radiomic studies of the placenta are based on T2WI images of MR. In the future, we will add more functional MR sequences to the artificial intelligence study to optimize the model. Third, ultrasound is important for the diagnosis of PAS and its complications; we will combine MR and ultrasound to optimize models. In addition, our model is based on pregnant women suspected of PAS, but all patients in our external validation group have been confirmed to have PAS. In the future, we will separately use pregnant women suspected of PAS by ultrasound but without PAS after surgery as external validation. Thus, in this way, we can verify the universality of our research. Conclusion This study provides a preoperative and noninvasive method for predicting PPH in high-risk PAS patients. In particular, the nomogram constructed by incorporating radiomics, intraplacental dark T2 bands, myometrial thinning/disruption, increased placental vascularity and loss of retroplacental T2-hypointense line can guide obstetricians in making reasonable preoperative treatment plans for high-risk PAS patients. Declarations Competing interests The authors declare no competing interests Funding This study was supported by Zhejiang Medical and Health Science Project (2022ZH088 to ZF. Wang) and(2022PY030 to HY.Wang). Author Contribution Chenying Lu and Zhihan Yan organized the research team and developed a research plan. Yumin Hu and Zufei Wang conducted statistical analysis and wrote the main manuscript text. Yechao Huang and Xia Li prepared figures. Bo Chen and Di Shen collected clinical data. Zhangwei Zhou and Haiyong Wang collected image data . Acknowledgement Our study is a retrospective study, therefore there is no clinical registration. Availability of data and materials Our study was a retrospective study, therefore there was no clinical registration.The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. References Linde LE, Rasmussen S, Moster D, Kessler J, Baghestan E, Gissler M, Ebbing C. Risk factors and recurrence of cause-specific postpartum hemorrhage: A population-based study. PLoS ONE. 2022;17(10):e0275879. Parant O, Guerby P, Bayoumeu F. Obstetric and anesthetic specificities in the management of a postpartum hemorrhage (PPH) associated with cesarean section. J De Gynecologie Obstetrique Et Biol De La Reprod. 2014;43(10):1104–22. Gallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, Williams MJ, Diaz V, Pasquale J, Chamillard M et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis. Cochrane Database Syst Reviews 2018(12). Thams AB, Larsen MH, Rasmussen SC, Jeppegaard M, Krebs L. Incidence of postpartum hemorrhage and risk factors for recurrence in the subsequent pregnancy. Arch Gynecol Obstet. 2023;307(4):1217–24. Ngwenya S. Postpartum hemorrhage: incidence, risk factors, and outcomes in a low-resource setting. Int J Womens Health. 2016;8:647–50. Feduniw S, Warzecha D, Szymusik I, Wielgos M. Epidemiology, prevention and management of early postpartum hemorrhage - a systematic review. Ginekologia polska. 2020;91(1):38–44. Zhang J, Xu H, Xin Y, Zhang C, Liu Z, Han X, Liu Q, Li Y, Huang Z. Assessment of the massive hemorrhage in placenta accreta spectrum with magnetic resonance imaging. Exp Ther Med. 2020;19(3):2367–76. Wu Q, Yao K, Liu Z, Li L, Zhao X, Wang S, Shang H, Lin Y, Wen Z, Zhang X, et al. Radiomics analysis of placenta on T2WI facilitates prediction of postpartum haemorrhage: A multicentre study. EBioMedicine. 2019;50:355–65. Bihan L, Nowak E, Anouilh F, Tremouilhac C, Merviel P, Tromeur C, Robin S, Drugmanne G, Le Roux L, Couturaud F et al. Development and Validation of a Predictive Tool for Postpartum Hemorrhage after Vaginal Delivery: A Prospective Cohort Study. Biology (Basel) 2022, 12(1). Hu Y, Weng Q, Xia H, Chen T, Kong C, Chen W, Pang P, Xu M, Lu C, Ji J. A radiomic nomogram based on arterial phase of CT for differential diagnosis of ovarian cancer. Abdom Radiol (NY). 2021;46(6):2384–92. Chen E, Mar WA, Horowitz JM, Allen A, Jha P, Cantrell DR, Cai K. Texture analysis of placental MRI: can it aid in the prenatal diagnosis of placenta accreta spectrum? Abdom Radiol. 2019;44(9):3175–84. Liu C, Xu Y, Li J, Guan Z, Liu C, He F. Development and validation of a predictive model for severe postpartum hemorrhage in women undergoing vaginal delivery: A retrospective cohort study. Int J Gynaecol Obstet. 2022;157(2):353–8. Romeo V, Ricciardi C, Cuocolo R, Stanzione A, Verde F, Sarno L, Improta G, Mainenti PP, D'Armiento M, Brunetti A, et al. Machine learning analysis of MRI-derived texture features to predict placenta accreta spectrum in patients with placenta previa. Magn Reson Imaging. 2019;64:71–6. Zhang Y, Hu M, Wen X, Huang Y, Luo R, Chen J. MRI-based radiomics nomogram in patients with high-risk placenta accreta spectrum: can it aid in the prenatal diagnosis of intraoperative blood loss? Abdominal radiology (New York) 2023. Song Z, Wang P, Zou L, Zhou Y, Wang X, Liu T, Zhang D. Enhancing postpartum hemorrhage prediction in pernicious placenta previa: a comparative study of magnetic resonance imaging and ultrasound nomogram. Front Physiol. 2023;14:1177795. Jha P, Pōder L, Bourgioti C, Bharwani N, Lewis S, Kamath A, Nougaret S, Soyer P, Weston M, Castillo RP, et al. Society of Abdominal Radiology (SAR) and European Society of Urogenital Radiology (ESUR) joint consensus statement for MR imaging of placenta accreta spectrum disorders. Eur Radiol. 2020;30(5):2604–15. Ding Y, Zhang C, Wu W, Pu J, Zhao X, Zhang H, Zhao L, Schoenhagen P, Liu S, Ma X. A radiomics model based on aortic computed tomography angiography: the impact on predicting the prognosis of patients with aortic intramural hematoma (IMH). Quant Imaging Med Surg. 2023;13(2):598–609. Allen L, Jauniaux E, Hobson S, Papillon-Smith J, Belfort MA. FIGO consensus guidelines on placenta accreta spectrum disorders: Nonconservative surgical management. Int J Gynaecol Obstet. 2018;140(3):281–90. Shamshirsaz AA, Fox KA, Salmanian B, Diaz-Arrastia CR, Lee W, Baker BW, Ballas J, Chen Q, Van Veen TR, Javadian P, et al. Maternal morbidity in patients with morbidly adherent placenta treated with and without a standardized multidisciplinary approach. Am J Obstet Gynecol. 2015;212(2):e218211–219. Silver RM, Fox KA, Barton JR, Abuhamad AZ, Simhan H, Huls CK, Belfort MA, Wright JD. Center of excellence for placenta accreta. Am J Obstet Gynecol. 2015;212(5):561–8. Peng L, Zhang X, Liu J, Liu Y, Huang J, Chen J, Su Y, Yang Z, Song T. MRI-radiomics-clinical-based nomogram for prenatal prediction of the placenta accreta spectrum disorders. Eur Radiol. 2022;32(11):7532–43. Bartels HC, O'Doherty J, Wolsztynski E, Brophy DP, MacDermott R, Atallah D, Saliba S, Young C, Downey P, Donnelly J, et al. Radiomics-based prediction of FIGO grade for placenta accreta spectrum. Eur Radiol Exp. 2023;7(1):54. Do QN, Lewis MA, Xi Y, Madhuranthakam AJ, Happe SK, Dashe JS, Lenkinski RE, Khan A, Twickler DM. MRI of the Placenta Accreta Spectrum (PAS) Disorder: Radiomics Analysis Correlates With Surgical and Pathological Outcome. J Magn Reson Imaging. 2020;51(3):936–46. Zhang Y, Hu M, Wen X, Huang Y, Luo R, Chen J. MRI-based radiomics nomogram in patients with high-risk placenta accreta spectrum: can it aid in the prenatal diagnosis of intraoperative blood loss? Abdom Radiol (NY). 2023;48(3):1107–18. Matsunaga S, Masuko H, Takai Y, Kanayama N, Seki H. Fibrinogen may aid in the early differentiation between amniotic fluid embolism and postpartum haemorrhage: a retrospective chart review. Springer Science and Business Media LLC 2021(1). Muñoz M, Stensballe J, Ducloy-Bouthors AS, Bonnet MP, De Robertis E, Fornet I, Goffinet F, Hofer S, Holzgreve W, Manrique S, et al. Patient blood management in obstetrics: prevention and treatment of postpartum haemorrhage. A NATA consensus statement. Blood Transfus. 2019;17(2):112–36. Hu Y, Wang Y, Weng Q, Wu X, Xia S, Wang H, Cheng X, Mao C, Li X, Zhou Z, et al. Intraplacental T2-hypointense bands may help predict placental invasion depth and postpartum hemorrhage in placenta accrete spectrum disorders in high-risk gravid patients. Magn Reson Imaging. 2022;94:73–9. Chen T, Xu XQ, Shi HB, Yang ZQ, Zhou X, Pan Y. Conventional MRI features for predicting the clinical outcome of patients with invasive placenta. Diagn Interv Radiol. 2017;23(3):173–9. Bourgioti C, Konstantinidou AE, Zafeiropoulou K, Antoniou A, Fotopoulos S, Theodora M, Daskalakis G, Nikolaidou ME, Tzavara C, Letsika A, et al. Intraplacental Fetal Vessel Diameter May Help Predict for Placental Invasiveness in Pregnant Women at High Risk for Placenta Accreta Spectrum Disorders. Radiology. 2021;298(2):403–12. Patel-Lippmann KK, Planz VB, Phillips CH, Ohlendorf JM, Zuckerwise LC, Moshiri M. Placenta Accreta Spectrum Disorders: Update and Pictorial Review of the SAR-ESUR Joint Consensus Statement for MRI. Radiographics. 2023;43(5):e220090. Tables Table 1 Patient clinical characteristics in the training and validation cohorts. Characteristics Training Cohort （ n ＝ 105 ） Independent validation cohort （ n=28 ） External validation cohort （ n=41 ） PPH( ＋ ) n=38 PPH （ - ） n=67 P value PPH( ＋ ) n=12 PPH （ - ） n ＝ 16 P value PPH( ＋ ) n=20 PPH （ - ） n=21 P value Maternal age, years (Mean ± SD) 34.50 (32.00-38.00) 32.00 (29-38) 0.108 32.50 (28.5-35.5) 33.50 (29.25-36.75) 0.701 34.50 (29.25-38) 34.00 （ 28.5-37 ） 0.647 Gestational age at MRI, weeks (Mean ±SD ) 34.00 (33-36) 34.00 (32-36) 0.985 34.00 (32-34.75) 34.00 (32.23-36.00) 0.673 33.00 （ 32-33.75 ） 33.00 （ 32-35 ） 0.261 Gestational age at CD , weeks (Mean ± SD) 36.00 (35-37 ） 37.00 (36-37) 0.037* 36 (35.00-37.00) 35.50 (34.25-36.75) 0.433 36.00 （ 34.25-36 ） 36.00 （ 35-37 ） 0.425 Previous CD history Negative Positive 12 26 37 30 0.020* 3 9 5 11 1.000 11 9 12 9 0.890 Number of Previous CD (Median,range) 1(0-1) 0(0-1) 0.01* 1(0.25-1) 1(0-1) 0.688 0 （ 0-1.75 ） 0 （ 0-1 ） 0.511 Abortion history Negative Positive 10 28 18 49 0.951 2 10 1 15 0.560 2 18 6 15 0.134 Number of Previous Abortion (Median,range) 1(0-2) 1(0-2) 0.995 1(1-2) 1.5(1-3) 0.159 2 （ 1-2 ） 2 （ 0-3 ） 0.925 Subtype of placenta previa Low‑lying Marginal Partial Complete 1 1 1 35 11 3 4 49 0.091 2 10 0 0 0 13 2 1 0.218 0 3 1 16 1 4 0 16 1.000 Depth of PAS invasiveness (intraoperative and/or histologic) No Accreta Increta Percreta 6 5 19 8 28 23 16 0 ＜ 0.001* 3 3 4 2 3 6 7 0 0.387 0 10 9 1 0 18 3 0 0.026 Blood transfusion Negative Positive 3 28 64 3 ＜ 0.001* 2 10 13 3 0.002* 4 16 19 2 0.000* Preoperative hemoglobin level (g/L) 109.66 ± 11.88 111.91 ± 9.81 0.298 102.17 ± 20.25 110.88 ± 11.88 0.165 111.45 ± 12.23 114.67 ± 14.31 0.445 Postoperative hemoglobin levels(g/L) 87.53 ± 14.19 101.42 ± 12.67 0.001* 92.83 ± 24.01 98.63 ± 16.95 0.460 103.85 ± 13.58 109.52 ± 15.49 0.221 Preoperative platelet count (×109) 177.63 ± 54.27 204.33 ± 64.30 0.033* 217.00 ± 50.58 184.25 ± 35.19 0.053 229.71 ± 76.86 211.40 ± 52.83 0.382 Postoperative platelet count (×109) 149.73 ± 60.59 202.91 ± 67.46 ＜ 0.001* 181.75 ± 54.42 195.75 ± 51.59 0.494 169.90 ± 40.55 228.19 ± 71.02 0.003 Preoperative PT 10.97 ± 1.51 10.80 ± 0.82 0.372 10.81 ± 0.92 10.72 ± 0.79 0.784 12.87 ± 1.37 12.50 ± 0.58 0.263 Preoperative APTT 27.11 ± 3.20 26.97 ± 2.84 0.820 26.75 ± 2.91 26.30 ± 1.86 0.622 33.96 ± 4.04 33.51 ± 3.85 0.716 Preoperative level of D-Dimer 5.59 ± 17.05 1.90 ± 1.26 0.082 3.54 ± 2.70 1.93 ± 1.64 0.061 3.93 ± 6.05 3.16 ± 4.30 0.703 Preoperative level of fibrinogen 4.170 ± 1.008 4.109 ± 0.902 0.752 3.79 ± 0.83 4.10 ± 0.92 0.371 4.28 ± 1.18 4.52 ± 1.01 0.478 CD cesarean delivery, SD standard deviation, PAS placenta accrete spectrum,PT prothrombin time ,APTT activated partial thromboplastin time,INR international normalized ratio, D-Dime Table 2 Predictive performance of MRI morphological features for PPH in training, validation, and external validation cohorts Cohort MRI morphological findings AUC Sensitivity Specificity Accuracy PPV NPV p value Training cohort Intraplacental dark T2 bands 0.689 (0.592-0.776) 64.18 (51.5 - 75.5) 73.68 (56.9 - 86.6) 67.619 (57.789-76.426) 81.132 (71.040-88.287) 53.846 (44.560-62.873) 0.0001* Placenta bulge 0.645 (0.545- 0.736) 100 (94.60-100.0) 28.95 (15.4 - 45.9) 74.038 (64.519-82.143) 70.968 (66.616-74.965) 100 (71.509-100) 0.0001* Loss of retroplacental T2-hypointense line 0.732 (0.636 - 0.813) 91.045 (81.520-96.642) 55.263 (38.299-71.376) 78.095 (68.967-85.579) 78.205 (71.430-83.739) 77.778 (60.765-88.776) ＜ 0.0001* Myometrial thinning/disruption 0.581 (0.481 -0.677) 92.537 (83.437-97.533) 23.684 (11.444-40.241) 67.619 (57.789-76.426) 68.132 (63.879-72.103) 64.286 (39.403-83.285) 0.0352* Bladder tenting sign 0.539 (0.439-0.637) 100 (94.643-100) 7.895 (1.659-21.377) 66.667 (56.799-75.567) 65.686 (63.559-67.753) 100 (29.240-46.145) 0.503 Increased placental vascularity 0.711 (0.614 -0.795) 89.552 (79.651-95.695) 52.632 (35.818-69.019) 76.190 (66.890-83.961) 76.923 (70.245-82.476) 74.074 (57.117-85.972) 0.000* Abnormal vascularization of the placental bed 0.561 (0.460-0.658) 91.045 (81.520-96.642) 23.684 (11.444-40.241) 66.667 (56.799-75.567) 67.778 (63.441-71.82) 60.000 (36.638-79.555) 0.1439 Independent validation cohort Intraplacental dark T2 bands 0.719 (0.518 - 0.871) 75.00 (42.8 - 94.5) 68.75 (41.3 - 89.0) 71.429 (51.333-86.776) 64.3 (44.8 - 80.0) 78.6 (56.6 - 91.2) 0.0135* Placenta bulge 0.594 (0.393-0.774) 25.00 (5.5 - 57.2) 93.75 (69.8 - 99.8) 64.286 (44.065-81.359) 75.0 (26.2-96.2) 62.5 (54.0 - 70.3) 0.1952 Loss of retroplacental T2-hypointense line 0.646 (0.444 - 0.816) 41.67 (15.2 - 72.3) 87.50 (61.7 - 98.4) 67.857 (47.648-84.122) 71.4 (36.8 - 91.5) 66.7 (54.5 - 77.0) 0.0889 Myometrial thinning/disruption 0.646 (0.444 - 0.816) 41.67 (15.2 - 72.3) 87.50 (61.7 - 98.4) 67.857 (47.648-84.122) 71.4 (36.8 - 91.5) 66.7 (54.5 - 77.0) 0.0889 Bladder tenting sign 0.583 (0.383-0.765) 8.33 (0.2 - 38.5) 100 (79.4 - 100.0) 42.857 (24.462-62.821) 100 （ 90.8-100 ） 42.9 (42.9 - 42.9) 0.3173 Increased placental vascularity 0.604 (0.403-0.783 33.33 (9.9 - 65.1) 87.50 (61.7-98.4) 64.286 (44.065-81.359) 66.7 (30.4-90.2) 63.6 (53.0-73.1) 0.2090 Abnormal vascularization of the placental bed 0.604 (0.403 - 0.783) 33.33 (9.900- 65.100) 87.50 (61.700-98.400) 64.286 (44.065-81.359) 66 (30.40 - 90.20) 63.6 (53.00 - 73.10) 0.2090 External validation cohort Intraplacental dark T2 bands 0.777 (0.620 -0.892) 65.00 (40.8 - 84.6) 90.48 (69.6 - 98.8) 63.415 （ 50.895-91.343 ） 86.7 (62.6 - 96.2) 73.1 (59.5 - 83.4) <0.0001* Placenta bulge 0.604 (0.439 - 0.753) 35.00 (15.4 - 59.2) 85.71 (63.7 - 97.0) 60.976 （ 44.505-75.799 ） 70.0 (41.1 - 88.6) 58.1 (49.0 - 66.6) 0.1236 Loss of retroplacental T2-hypointense line 0.631 (0.466 - 0.776) 50.00 (27.2 - 72.8) 76.19 (52.8 - 91.8) 63.415 （ 46.936-77.877 ） 66.7 (45.3 - 82.8) 61.5 (49.3 - 72.5) 0.0790 Myometrial thinning/disruption 0.702 (0.539- 0.835) 50.00 (27.196 - 72.804) 90.476 (69.623 - 98.825) 70.732 （ 54.463-83.870 ） 83.333 (55.483 - 95.251) 65.517 (54.541 - 75.05) 0.0022* Bladder tenting sign 0.524 (0.362 - 0.682) 100.00 (83.2 - 100.0) 4.76 (0.1 - 23.8) 48.780 （ 32.878-64.866 ） 50.0 (47.6 - 52.4) 100 (90.8-100) 0.3173 Increased placental vascularity 0.777 (0.620 - 0.892) 65.00 (40.8 - 84.6) 90.48 (69.6 - 98.8) 78.049 （ 62.386-89.439 ） 86.7 (62.6 - 96.2) 73.1 (59.5 - 83.4) <0.0001* Abnormal vascularization of the placental bed 0.602 (0.438 - 0.752) 30.00 (11.9 - 54.3) 90.48 (69.6 - 98.8) 60.98 （ 44.505-75.799 ） 75.00 (40.6 - 92.9) 57.60 (49.7 - 65.1) 0.0985 Table 3 Models comparison auc auc 95ci specifictiy recall precision f-score accuracy training corhort Radiomic 0.929 (0.87, 0.987) 0.970 0.692 0.931 0.794 0.867 clinic 0.734 (0.631, 0.837) 0.970 0.513 0.909 0.656 0.800 combine 0.956 (0.91, 1.000) 0.985 0.846 0.971 0.904 0.933 independence validation corhort Radiomic 0.735 (0.547, 0.922) 0.857 0.357 0.714 0.476 0.607 clinic 0.584 (0.37, 0.798) 0.929 0.286 0.800 0.421 0.607 combine 0.781 (0.606, 0.955) 0.857 0.500 0.778 0.609 0.679 external validation corhort Radiomic 0.686 (0.522, 0.850) 0.810 0.400 0.665 0.500 0.610 clinic 0.694 (0.531, 0.857) 0.952 0.450 0.900 0.600 0.707 combine 0.702 (0.541, 0.864) 0.571 0.700 0.609 0.651 0.634 Additional Declarations No competing interests reported. Supplementary Files Supplementaltable.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4550980","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":317273459,"identity":"3eb60c03-5231-4238-ac21-2fadc69a4a53","order_by":0,"name":"Yumin Hu","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yumin","middleName":"","lastName":"Hu","suffix":""},{"id":317273462,"identity":"51169e2f-5768-4bc6-b47e-c29a27457ca7","order_by":1,"name":"Yechao Huang","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yechao","middleName":"","lastName":"Huang","suffix":""},{"id":317273468,"identity":"cf550258-208d-4d54-8554-bd4d86a9e04d","order_by":2,"name":"Bo Chen","email":"","orcid":"","institution":"The Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Bo","middleName":"","lastName":"Chen","suffix":""},{"id":317273470,"identity":"a96ded12-121a-4366-bab1-231ccd1b9ef8","order_by":3,"name":"Di Shen","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Di","middleName":"","lastName":"Shen","suffix":""},{"id":317273471,"identity":"a9314d5d-4206-48a6-80f7-182365c87334","order_by":4,"name":"Xia Li","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xia","middleName":"","lastName":"Li","suffix":""},{"id":317273473,"identity":"33e08ec3-cd5f-45a3-851d-6a8990f0482c","order_by":5,"name":"Zufei Wang","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zufei","middleName":"","lastName":"Wang","suffix":""},{"id":317273476,"identity":"1dcc1d5c-b1f1-4d5f-874d-bd11210aabae","order_by":6,"name":"Zhangwei Zhou","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhangwei","middleName":"","lastName":"Zhou","suffix":""},{"id":317273477,"identity":"4ea4c1dd-b50a-48f8-acd8-0af14b439eea","order_by":7,"name":"Haiyong Wang","email":"","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Haiyong","middleName":"","lastName":"Wang","suffix":""},{"id":317273478,"identity":"53c41d3e-8b9a-4e13-8140-a169c3cf7f77","order_by":8,"name":"Zhihan Yan","email":"","orcid":"","institution":"The Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhihan","middleName":"","lastName":"Yan","suffix":""},{"id":317273481,"identity":"a9246509-c466-4ab0-8be0-c4a470b570e8","order_by":9,"name":"Chenying Lu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6klEQVRIiWNgGAWjYBACPgYGNhDNw8DAfIBBAsQ8QEALG0ILWwJpWkC6DCA0QS0Syc8efNxxWMacf83nD5ZtDHJ8NxIYPxfg1ZJmbjjzzGEeyxlvNxhItjEYS95IYJaegVdLDps0b9thHoMbZzckALUkbriRwMbMQ0jLX7CWMw8OALXUE6eFEaTlfA9jA1BLggFBLTzPzCR729KBtrAZM0ickwB67GGzND4t/OzJzyR+tlnbG5w//PizRJmNPN/x5IOf8WmBgmYGBokEBmYJcGQyNhDWwMBQB7TvAAPjB2LUjoJRMApGwYgDAEUPRuOSJFbtAAAAAElFTkSuQmCC","orcid":"","institution":"The Fifth Affiliated Hospital of Wenzhou Medical University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Chenying","middleName":"","lastName":"Lu","suffix":""}],"badges":[],"createdAt":"2024-06-08 14:36:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4550980/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4550980/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":60447017,"identity":"9b65f07e-ebbe-445e-bc0d-7d594744a7ea","added_by":"auto","created_at":"2024-07-16 21:58:34","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":105126,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart for the recruitment of patients.\u003c/p\u003e","description":"","filename":"Slide1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/add6aad99a7d9cfa459a911c.jpg"},{"id":60447018,"identity":"10bd4247-09e6-453a-92f0-bce81d477308","added_by":"auto","created_at":"2024-07-16 21:58:34","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":160785,"visible":true,"origin":"","legend":"\u003cp\u003eThe radiomics workflow. This study included MR imaging segmentation, feature selection, model construction and evaluation. First, we selected placenta T2WI sagittal images to mask placenta and obtain VOI. The placental tissues were segmented to generate volume of interest (VOI) and then 987quantitative radiomic features were extracted from placenta VOI. Second, the least absolute shrinkage and selection operator (LASSO) was then used to select the features which were highly correlated with PPH. Based on the selected radiomic features, a radiomic signature was constructed using the support vector machine model, which is called radiomic model. Third, clinical factors were combined with radiomic signature to generate clinicoradiomic model. Model comparison was conducted to select the optimal model with best performance for postpartum haemorrhage prediction. Finally, the calibration and decision curve analysis were conducted to validate the performance and clinical value of the optimal model.\u003c/p\u003e","description":"","filename":"Slide2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/ac0f74d8c19c9fdf179673c1.jpg"},{"id":60447720,"identity":"47583d6e-c824-433f-9952-7d6bed62a09c","added_by":"auto","created_at":"2024-07-16 22:06:34","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":89511,"visible":true,"origin":"","legend":"\u003cp\u003eClinical and radiomic feature selection using the least absolute shrinkage and selection operator algorithm. (A) The variation of the clinical coefficients with log(λ) and the dotted line indicate the selected optimal log(λ) value (0.024) with min mean MSE. (B) The 4 clinical features with non-zero coefcients were obtained at the optimal λ value. (C) The variation of the radiomic coefficients with log(λ) and the dotted line indicate the selected optimal log(λ) (0.009) value with min mean MSE. (D) The 18 radiomic features with non-zero coefcients were obtained at the optimal λ value. (E)Heatmap of 18 radiomic features in the training cohort (n=105), with each row representing a feature and each column corresponding to one patient.\u003c/p\u003e","description":"","filename":"Slide3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/c9b0819c3dd213e83516f9b7.jpg"},{"id":60447015,"identity":"2d35ac45-c2b3-4963-84d5-c1288e68616f","added_by":"auto","created_at":"2024-07-16 21:58:34","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":70010,"visible":true,"origin":"","legend":"\u003cp\u003eWaterfall plot showing radiomic score (rad-score) for each patient in the (a) training cohort, (c) independent validation cohort, and (e) external validation cohort . Pregnant women with postpartum hemorrhage (PPH+) and without postpartum hemorrhage(PPH-) are marked with different colors ( purple: pregnant women with postpartum hemorrhage;green: pregnant women without postpartum hemorrhage). Rad-score in the PPH(+) group were signifcantly higher than those in the PPH(-) group in training(b) ,independent validation(d) and external validation(f) cohorts.\u003c/p\u003e","description":"","filename":"Slide4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/6bef54e6a988a8cca497ca70.jpg"},{"id":60447020,"identity":"a0c572d6-4c0b-4b0d-893e-2cade8b53d86","added_by":"auto","created_at":"2024-07-16 21:58:35","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":75975,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of ROC curves of different models. (a) - (c) are ROC curves of the models in training ,independent validation and external validation cohorts. Among the 3 models, the clinicoradiomic model resulted in best performance, with an AUC of 0.965 in the training cohort, an AUC of 0.781 in the independent validation cohort and an AUC of 0.702 in the external validation cohort.\u003c/p\u003e","description":"","filename":"Slide5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/1dac91d698bb44ec78793bfc.jpg"},{"id":60447019,"identity":"f862c1f2-ea22-4081-b763-8067ed778f98","added_by":"auto","created_at":"2024-07-16 21:58:34","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":70758,"visible":true,"origin":"","legend":"\u003cp\u003eThe establishment and performance of the clinicoradiomic model. (a) The developed nomogram based on clinicoradiomic model. (b-d) Calibration curves of the clinicoradiomic model generated from training ,independent validation and external validation cohorts. The goodness of fits of predicted probability from the clinicoradiomic model with the actual outcomes of postpartum haemorrhage (PPH) was assessed. The x-axis represented the probability of PPH calculated by the clinicoradiomic model while the y-axis represented the actual rate of PPH. Diagonal dotted line represented a perfect estimation by an ideal model, in which the estimated outcome perfectly corresponded to the actual outcome. Red line represented performance of the clinicoradiomic model, a closer alignment of which with the diagonal dotted line represented a better estimation. (d-f) Decision curve analysis for the clinicoradiomic model. The x-axis showed the threshold probability and y-axis measured the net benefit. The blue line represents the clinicoradiomic model. The grey line represented the assumption that all patients were PPH while the black line represented the assumption that no patients were PPH.\u003c/p\u003e","description":"","filename":"Slide6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/414c31d4a3efb43b896dea47.jpg"},{"id":60449033,"identity":"6912aadd-3e33-4b2d-b222-64cff98891e9","added_by":"auto","created_at":"2024-07-16 22:22:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3054740,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/5d7c3685-8b9a-41fa-82a3-486fc698df42.pdf"},{"id":60447014,"identity":"fee01db5-d950-47aa-87aa-9b5c43be38e6","added_by":"auto","created_at":"2024-07-16 21:58:34","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":13944,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaltable.docx","url":"https://assets-eu.researchsquare.com/files/rs-4550980/v1/bd068c2d7f88d9e085e4b7d7.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"MRI–radiomics–clinical–based nomogram for prediction postpartum hemorrhage in patients with suspected placenta accrete spectrum before cesarean section","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePostpartum hemorrhage (PPH) is the leading direct cause of maternal mortality worldwide[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].PPH is defined as 500 mL or more of blood loss from the genital tract following a normal vaginal delivery (NVD) or 1000 mL or more following a cesarean section within 24 hours of birth[\u003cspan additionalcitationids=\"CR3 CR4 CR5\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].Postpartum hemorrhage is the most common complication of Placenta Accrete Spectrum (PAS). It may lead to uncontrollable bleeding and threaten the lives of mother and baby[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Early prediction of PPH in suspected PAS patients would aid delivery planning, facilitate the preparation of blood products and transfusion requirements, as well as decrease maternal complications[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAt present, the prediction of postpartum hemorrhage in patients with suspected PAS includes laboratory indicators of coagulation function, clinical risk factors of PAS, and evaluation of ultrasound and magnetic resonance. However, these indicators are relatively scattered, and the prediction ability of postpartum hemorrhage is limited. Therefore, we need to build a model with predictive probability to predict postpartum hemorrhage. The development of a predictive model could optimize maternal outcomes by improving clinical recognition and timely treatment to avoid excessive blood loss in women with suspected PAS pregnancies [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRadiomics analysis quantifying high-dimensional tissue features that cannot be observed by the naked eye has shown great potential in precision medicine, aiming to extract mineable high-dimensional data from clinical images, which was followed by subsequent data analysis for decision support[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Although radiomic analysis is mostly applied to tumors, some studies found it to be a feasible tool for PAS diagnosis and even PAS severity assessment[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].However, there are not many studies on prenatal prediction of postpartum hemorrhage in pregnant women suspected of having PAS, which combined radiomic features with clinical, laboratory and MR morphological indicators in their studies[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this retrospective study, we aimed to extract the radiomic features of the placenta that correlated with PPH and at the same time, develop a nomogram for a predictive model that could optimize the maternal outcomes of patients with suspected PAS.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003e2.1 Patients\u003c/p\u003e \u003cp\u003e This retrospective study was approved by the Institutional Review Board of the Lishui Central Hospital (center 1) and the Second Affiliated Hospital and Yuying Children\u0026rsquo;s Hospital, Wenzhou Medical University (center 2), with written informed consent waived. Pregnant women suspicious of PAS clinically and on ultrasonic examination were collected from the electronic medical record system of Center 1 (n\u0026thinsp;=\u0026thinsp;155) from September 2016 to April 2022 and Center 2 (n\u0026thinsp;=\u0026thinsp;60) from July 2020 to January 2022. The inclusion criteria were (a) singleton pregnancies with intraoperative bleeding and bleeding volume recorded in postoperative 24-hour nursing records; (b) pregnancies that underwent MRI before cesarean operation; and (c) pregnancies that had complete pathologic and clinical information. The exclusion criteria were (a) patients with vaginal delivery (n\u0026thinsp;=\u0026thinsp;3 in center 1; n\u0026thinsp;=\u0026thinsp;4 in center 2) (b) incomplete clinical data (n\u0026thinsp;=\u0026thinsp;7 in center 1; n\u0026thinsp;=\u0026thinsp;8 in center 2) (c) poor image quality(n\u0026thinsp;=\u0026thinsp;8 in center 1; n\u0026thinsp;=\u0026thinsp;5 in center 2) (d) fatal congenital dysplasia (n\u0026thinsp;=\u0026thinsp;3 in center 1; n\u0026thinsp;=\u0026thinsp;2 in center 2).\u003c/p\u003e \u003cp\u003eFinally, there were 133 pregnant women (n\u0026thinsp;=\u0026thinsp;50 in the PPH(+) group; n\u0026thinsp;=\u0026thinsp;83 in the PPH(-) group) in center 1 and 41 pregnant women (n\u0026thinsp;=\u0026thinsp;20 in the PPH(+) group; n\u0026thinsp;=\u0026thinsp;21 in the PPH(-) group) in center 2 identified in our study. Pregnancies from Center 1 were randomized into a training (n\u0026thinsp;=\u0026thinsp;105) and an independent validation (n\u0026thinsp;=\u0026thinsp;28) cohort with a proportion of 8:2. Pregnancies from Center 2 were scheduled as the external validation cohort (n\u0026thinsp;=\u0026thinsp;41) (Fig.\u0026nbsp;1).\u003c/p\u003e \u003cp\u003e2.2. MRI examination\u003c/p\u003e \u003cp\u003eThe pelvic MR image acquisition of all pregnant women was performed on a 1.5-T MRI scanner (MAGENTOM Area; Siemens) in center 1 and a 1.5-T MRI scanner (MAGENTOM Avanto, Siemens) in center 2, both with a 6-channel abdominal phased-array surface coil. The sequence we used for this study is the sagittal FS-T2WI images. The parameters of MRI sequences in these two centers are presented in Supplemental Table\u0026nbsp;1.\u003c/p\u003e \u003cp\u003e2.3 Radiologic evaluation\u003c/p\u003e \u003cp\u003eTwo radiologists (CL and ZW with 16 and 21 years of experience in imaging diagnosis) evaluated the image findings separately on a RAD info RSVS system. Both readers were blinded to previous MRI interpretation, previous ultrasound reports, and the surgical and pathologic reports from delivery. Any disagreements were resolved by consensus. Seven MR features, according to the consensus statement of the Society of Abdominal Radiology (SAR) and the European Society of Urogenital Radiology (ESUR),were evaluated and recorded: intraplacental dark T2 bands, placenta bulge, loss of retroplacental T2-hypointense line, myometrial thinning/disruption, bladder tenting sign, increased placental vascularity, and abnormal vascularization of the placental bed[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e2.4 MR image segmentation and radiomic features extraction\u003c/p\u003e \u003cp\u003eThe DICOM of sagittal T2WI images, which is the optimal position for the uterus and placenta observation, was retrieved from PACS and uploaded to 3D slicer(\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.slicer.org\u003c/span\u003e\u003cspan address=\"https://www.slicer.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). All volume of interest (VOI) segmentation was conducted by a radiologist (CL) who had 16 years of imaging diagnosis experience and was re-checked by SX (15 years of experience); both were blind to the pathology reports. VOI segmentations were circled on the sagittal T2WI of the placenta. Quantitative radiomics features were extracted from VOIs on Pyradiomics(3.0.1). The extracted features were divided into three categories: shape-based features, first-order statistics,and texture features.\u003c/p\u003e \u003cp\u003e2.5 Radiomic and clinical features selection\u003c/p\u003e \u003cp\u003e987 radiomic features were extracted from the placenta volume of interest (VOI), including i) 14 shape features, iii) 203first order features, iiii) 770 texture features. To select the radiomic features highly correlated with PPH, we used least absolute shrinkage and selection operator (LASSO) regression with three-fold cross-validation. For the LASSO algorithm, L1 regularizer was used as the cost function, the error value of cross-validation was 5, and the maximum number of iterations was 1000.\u003c/p\u003e \u003cp\u003eThe clinical features include the basic data of prenatal patients, laboratory indicators and MR signs. The basic data of the patients included maternal age, gestational age at CD, previous CD history, number of previous abortions, number of abortions, presence of placenta previa, and times of prenatal bleeding. Preoperative laboratory indicators included hemoglobin level, platelet level, prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), D-Dimer, fibrinogen. MR signs included placenta bulge,loss of retroplacental T2-hypointense line,myometrial thinning/disruption,bladder tenting sign,increased placental vascularity,abnormal vascularization of the placental bed. So the LASSO method was also used to obtain the subset of predictors.\u003c/p\u003e \u003cp\u003e2.6 Radiomic and clinicoradiomic model building\u003c/p\u003e \u003cp\u003eAfter feature selection, we built the radiomic model using the support vector machine (SVM) algorithm with a radial basis kernel for risk prediction of PPH based on the selected radiomic features. The area under the receiver operator characteristic (ROC) curve (AUC) was used to validate the performance of the developed radiomics models in training, internal validation and external validation cohorts. The best predictive model was determined, and the probability output by the model was recorded as a radiomics score (Radscore).After that, the selected clinical predictor was added to the radiomic model to form a clinicoradiomic model[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e2.7 Development and validation of the nomogram\u003c/p\u003e \u003cp\u003eAfter model comparison, a nomogram for clinical use was built based on the model with the best discrimination ability for PPH. The calibration curves were used to evaluate the agreement between the observed results and predicted probability. The clinical utility of the nomogram was evaluated by quantifying the net benefit under different threshold probabilities by decision curve analysis. The work flow was shown in Fig.\u0026nbsp;2.\u003c/p\u003e \u003cp\u003e2.8 Statistical analysis\u003c/p\u003e \u003cp\u003eStatistical analysis was performed using SPSS (SPSS 27.0; IBM Corp), MedCalc and R Programming Language (version 3.5.5; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.Rproject.org\u003c/span\u003e\u003cspan address=\"http://www.Rproject.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The SPSS software was used to perform the variable comparison between different groups or cohorts. Continuous variables were compared by the Student t test or Mann\u0026ndash;Whitney U test between PPH(་) and PPH(-) groups. Categorical variables were compared using the 2\u0026times;2 or R\u0026times;C χ2 test or Fisher exact test. Predictive performance(AUC, Sensitivity, Specificity, Accuracy, PPV, NPV, P value) of MRI morphological features for PPH were calculated by MedCalc. The R Programming Language was applied to conduct the LASSO regression analysis, ROC analysis (with the area under the curve [AUC] value calculated), calibration curve analysis (with the Hosmer\u0026ndash;Lemeshow test), and decision curve analysis. The AUC values among different models were compared using the DeLong method. A p value less than 0.05 was set as the statistical significance threshold. Nomogram development and calibration plots were performed using the \u0026ldquo;rms\u0026rdquo; package. The decision curve analysis was performed using the \u0026ldquo;dca.R.\u0026rdquo; package.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Clinical characteristics and MR features of patients\u003c/h2\u003e\n \u003cp\u003eThe clinical characteristics before and after the cesarean section of all pregnant women were summarized in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. There were statistical differences in gestational age at CD (P\u0026thinsp;=\u0026thinsp;0.037), history of cesarean section (P\u0026thinsp;=\u0026thinsp;0.020), number of cesarean section (P\u0026thinsp;=\u0026thinsp;0.010), depth of PAS invasiveness (P\u0026lt;0.001), whether postpartum blood transfusion performed (P\u0026lt;0.001), preoperative platelet count (P\u0026thinsp;=\u0026thinsp;0.033), postoperative hemoglobin levels (P\u0026thinsp;=\u0026thinsp;0.001) and platelet count (P\u0026lt;0.001) between the groups with and without PPH in the training group. There were statistical differences in whether postpartum blood transfusion was performed between the groups with and without PPH in independent validation cohort (P\u0026thinsp;=\u0026thinsp;0.002) and the external validation cohort (P\u0026thinsp;=\u0026thinsp;0.000).\u003c/p\u003e\n \u003cp\u003eThe diagnostic performance of MRI morphological findings for PAS in the training, independent validation, and external validation cohorts were shown in Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, respectively. Among seven MRI morphological findings, five MRI morphological findings were significantly different between PPH(+) and PPH(-) pregnancies in the training cohort(Intraplacental dark T2 bands P\u0026thinsp;=\u0026thinsp;0.0001,Placenta bulge P\u0026thinsp;=\u0026thinsp;0.0001,Loss of retroplacental T2-hypointense line P\u0026lt;0.0001,Myometrial thinning/disruption 0.0352,Increased placental vascularity P\u0026thinsp;=\u0026thinsp;0.000),one MRI morphological finding significantly different between PPH(+) and PPH(-) pregnancies in the independent validation cohort(Intraplacental dark T2 bands P\u0026thinsp;=\u0026thinsp;0.0135),and three MRI morphological findings significantly different between PPH(+) and PPH(-) pregnancies in the external validation cohort(Intraplacental dark T2 bands P\u0026thinsp;\u0026lt;\u0026thinsp;0.0001,Myometrial thinning/disruption P\u0026thinsp;=\u0026thinsp;0.0022,Increased placental vascularity\u0026thinsp;\u0026lt;\u0026thinsp;0.0001).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 The clinical and selected radiomic features\u003c/h2\u003e\n \u003cp\u003eFour clinical features were selected by LASSO from 20 clinical features(Figure3a,b), all of which are imaging features, including intraplacental dark T2 bands, myometrial thinning/disruption, increased placental vascularity and loss of retroplacental T2-hypointense line. In total, 987 radiomic features were extracted from each VOI of the placenta on T2WI. Among them, 18 features were selected using the LASSO regression algorithm.༈Figure3c-e༉.The 18 radiomic features can be seen in Supplemental table 2.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3 Models construction and comparison\u003c/h2\u003e\n \u003cp\u003eA radiomic model was constructed using the SVM. The AUC of radiomic signature from the training cohort was 0.929 (95% CI, 0.870, 0.987), from the independent validation cohort were 0.735 (95% CI,, 0.547, 0.922) and from the external validation cohort were 0.686(95% CI, 0.522, 0.850). The SVM model was used to calculate the radiomics signature Rad-score. The Rad-score for each patient presented as a waterfall plot demonstrated significant differences between PPH (+) and PPH (-) in training, internal validation and external validation cohorts (Fig.\u0026nbsp;4a,c,e). PPH(+) and PPH(-) showed significantly different rad-scores in training, internal validation, and external validation cohorts (Fig.\u0026nbsp;4b,d,f).\u003c/p\u003e\n \u003cp\u003eThe clinicoradiomic model, developed by incorporating radiomic signature with clinical factors, resulted in an AUC of 0.956 (95% CI, 0.911) in the training cohort, 0.781 (95% CI, 0.606 0.955) in the independent validation cohort, 0.702(95% CI, 0.541 0.864) in the external validation cohort respectively for PPH prediction. The performance of radiomic signature, clinical and clinicoradiomic models were shown in Fig. 5 and Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\n \u003cp\u003eThe performance of the clinicoradiomic models were better than that of the clinical models and radiomic models in all three cohorts.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4 Construction and validation of clinicoradiomic nomogram\u003c/h2\u003e\n \u003cp\u003eThe nomogram was defined based on the clinicoradiomic model Fig.\u0026nbsp;6a, which showed the best discrimination ability among 3 models. The calibration curve, together with the H-L test, was used to estimate the consistency between the probability of PPH predicted by the clinicoradiomic model and actual outcomes.As shown in Fig.\u0026nbsp;6b\u0026ndash;d, the predicted probability of PPH in the training, independent validation and external validation cohorts respectively match the anticipated probability of PPH.\u003c/p\u003e\n \u003cp\u003eThe decision curve evaluated the performance of the clinicoradiomic model in terms of clinical application, hence, reflecting its clinical usefulness. DCA indicated that the farther the decision curve was from the two extreme curves, the higher the clinical decision net benefit. In this study, the net benefit of the clinicoradiomic model were higher than that of the clinical model(training cohort Z\u0026thinsp;=\u0026thinsp;4.0234,P\u0026thinsp;=\u0026thinsp;5.73\u0026times;10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e, independent validation cohort Z\u0026thinsp;=\u0026thinsp;1.4005,P\u0026thinsp;=\u0026thinsp;0.1614, external validation cohort Z\u0026thinsp;=\u0026thinsp;0.093867,P\u0026thinsp;=\u0026thinsp;0.9252) and the radiomic model(training cohort Z=-1.3121,P\u0026thinsp;=\u0026thinsp;0.1895, independent validation cohort Z=-0.79029,P\u0026thinsp;=\u0026thinsp;0.4294, external validation cohort Z=-0.13376,P\u0026thinsp;=\u0026thinsp;0.8936) (Fig.\u0026nbsp;6e-g), indicating good discrimination in training, independent validation cohort and external validation cohort.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, radiomic features based on sagittal FS-T2WI placental MR images were selected and used to construct a radiomic model. In order to facilitate visualization, clinical application, and decision-making, we further developed and validated a nomogram that combines clinical and radiomics features for antenatal prediction of postpartum hemorrhage in patients with suspected placenta accrete spectrum. The results showed that the nomogram was superior to the clinical and radiomics models in preoperatively predicting PPH.\u003c/p\u003e \u003cp\u003eThe precise prenatal diagnosis and prediction of postpartum complications of PAS are increasingly receiving attention from obstetricians[\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].In recent years, some studies have applied artificial intelligence to the study of PAS[\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] and its postpartum complications[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].There are several differences from previous studies. Firstly, our study had an external validation, which ensured the effectiveness of the training model.Secondly, the patients included in our study included all patients suspected to have PAS by ultrasound, including those confirmed by postoperative pathology, those diagnosed by obstetricians during surgery, and those without PAS. Because we cannot accurately diagnose PAS in pregnant women before surgery, our grouping and constructed model are more in line with clinical needs. In our study, we extracted 18 radiomic features from the training group, and the SVM model based on the 18 radiomic features had good performance in the training cohort(auc\u0026thinsp;=\u0026thinsp;0.929), independent validation cohort༈auc\u0026thinsp;=\u0026thinsp;0.735༉and external validation cohort༈auc\u0026thinsp;=\u0026thinsp;0.686༉. The SVM model we constructed was consistent with previous studies which used in another article on prediction of postpartum haemorrhage. This further confirmed that the SVM model had a good predictive effect on postpartum hemorrhage in the placenta accreta spectrum[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eUnlike previous studies, clinical factors related to postpartum hemorrhage, such as prenatal coagulation function, pregnancy history, and prenatal placental MR features, were used together to screen for the best clinical features. And we found that in both baseline table calculations and lasso feature screening, prenatal coagulation function indicators were not statistically significant predictors of postpartum hemorrhage. This may be due to the fact that the prenatal coagulation function of PAS is mostly normal, and abnormalities in coagulation function often occur after PPH during cesarean section[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].Finally, four MR features (intraplacental dark T2 bands, increased placental vascularity, myometrial thinning/disruption, loss of retroplacental T2-hypointense line)we screened from 20 clinical features by lasso regression were associated with PPH. And these four MR features were important features associated with postpartum complications in previous studies. Previous studies suggested that intraplacental dark T2 bands bands may help predict placental invasion depth and postpartum hemorrhage in placenta accrete spectrum disorders in high-risk gravid patients[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. In addition, Bourgioti\u0026rsquo;s study suggested that increased placental vascularities were valuable for the prediction of postpartum hemorrhage[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Myometrial thinning/disruption and loss of retroplacental T2-hypointense line were also important signs reflecting the severity of PAS[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].The AUC of clinical model with the combination of these four MR features was 0.734 in training cohort, 0.584 in independent validation cohort,0.694 in external validation cohort.\u003c/p\u003e \u003cp\u003eAfter combining radiomic models with clinical models, the effectiveness of the combined models was significantly improved. The AUC of combined model in training cohort was 0.956, 0.781 in independent validation cohort, 0.702 in external validation cohort. Although it is hard to interpret the specific underlying meaning of radiomic features, radiomic analysis combined with clinical analysis is a useful and stable tool for PPH prediction. The stable performance in different centers also highlights that radiomics analysis can overcome data differences between multiple center.\u003c/p\u003e \u003cp\u003eAlthough our results are promising, there are several limitations to our study. First of all, the number of patients was relatively small. We will increase the sample size and the number of external verifications from other institutions. Secondly, the radiomic studies of the placenta are based on T2WI images of MR. In the future, we will add more functional MR sequences to the artificial intelligence study to optimize the model. Third, ultrasound is important for the diagnosis of PAS and its complications; we will combine MR and ultrasound to optimize models. In addition, our model is based on pregnant women suspected of PAS, but all patients in our external validation group have been confirmed to have PAS. In the future, we will separately use pregnant women suspected of PAS by ultrasound but without PAS after surgery as external validation. Thus, in this way, we can verify the universality of our research.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study provides a preoperative and noninvasive method for predicting PPH in high-risk PAS patients. In particular, the nomogram constructed by incorporating radiomics, intraplacental dark T2 bands, myometrial thinning/disruption, increased placental vascularity and loss of retroplacental T2-hypointense line can guide obstetricians in making reasonable preoperative treatment plans for high-risk PAS patients.\u003c/p\u003e "},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis study was supported by Zhejiang Medical and Health Science Project (2022ZH088 to ZF. Wang) and(2022PY030 to HY.Wang).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eChenying Lu and Zhihan Yan organized the research team and developed a research plan. Yumin Hu and Zufei Wang conducted statistical analysis and wrote the main manuscript text. Yechao Huang and Xia Li prepared figures. Bo Chen and Di Shen collected clinical data. Zhangwei Zhou and Haiyong Wang collected image data .\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eOur study is a retrospective study, therefore there is no clinical registration.\u003c/p\u003e\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e \u003cp\u003eOur study was a retrospective study, therefore there was no clinical registration.The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLinde LE, Rasmussen S, Moster D, Kessler J, Baghestan E, Gissler M, Ebbing C. Risk factors and recurrence of cause-specific postpartum hemorrhage: A population-based study. PLoS ONE. 2022;17(10):e0275879.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParant O, Guerby P, Bayoumeu F. Obstetric and anesthetic specificities in the management of a postpartum hemorrhage (PPH) associated with cesarean section. J De Gynecologie Obstetrique Et Biol De La Reprod. 2014;43(10):1104\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGallos ID, Papadopoulou A, Man R, Athanasopoulos N, Tobias A, Price MJ, Williams MJ, Diaz V, Pasquale J, Chamillard M et al. Uterotonic agents for preventing postpartum haemorrhage: a network meta-analysis. Cochrane Database Syst Reviews 2018(12).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThams AB, Larsen MH, Rasmussen SC, Jeppegaard M, Krebs L. Incidence of postpartum hemorrhage and risk factors for recurrence in the subsequent pregnancy. Arch Gynecol Obstet. 2023;307(4):1217\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNgwenya S. Postpartum hemorrhage: incidence, risk factors, and outcomes in a low-resource setting. Int J Womens Health. 2016;8:647\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeduniw S, Warzecha D, Szymusik I, Wielgos M. Epidemiology, prevention and management of early postpartum hemorrhage - a systematic review. Ginekologia polska. 2020;91(1):38\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Xu H, Xin Y, Zhang C, Liu Z, Han X, Liu Q, Li Y, Huang Z. Assessment of the massive hemorrhage in placenta accreta spectrum with magnetic resonance imaging. Exp Ther Med. 2020;19(3):2367\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu Q, Yao K, Liu Z, Li L, Zhao X, Wang S, Shang H, Lin Y, Wen Z, Zhang X, et al. Radiomics analysis of placenta on T2WI facilitates prediction of postpartum haemorrhage: A multicentre study. EBioMedicine. 2019;50:355\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBihan L, Nowak E, Anouilh F, Tremouilhac C, Merviel P, Tromeur C, Robin S, Drugmanne G, Le Roux L, Couturaud F et al. Development and Validation of a Predictive Tool for Postpartum Hemorrhage after Vaginal Delivery: A Prospective Cohort Study. Biology (Basel) 2022, 12(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu Y, Weng Q, Xia H, Chen T, Kong C, Chen W, Pang P, Xu M, Lu C, Ji J. A radiomic nomogram based on arterial phase of CT for differential diagnosis of ovarian cancer. Abdom Radiol (NY). 2021;46(6):2384\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen E, Mar WA, Horowitz JM, Allen A, Jha P, Cantrell DR, Cai K. Texture analysis of placental MRI: can it aid in the prenatal diagnosis of placenta accreta spectrum? Abdom Radiol. 2019;44(9):3175\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu C, Xu Y, Li J, Guan Z, Liu C, He F. Development and validation of a predictive model for severe postpartum hemorrhage in women undergoing vaginal delivery: A retrospective cohort study. Int J Gynaecol Obstet. 2022;157(2):353\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRomeo V, Ricciardi C, Cuocolo R, Stanzione A, Verde F, Sarno L, Improta G, Mainenti PP, D'Armiento M, Brunetti A, et al. Machine learning analysis of MRI-derived texture features to predict placenta accreta spectrum in patients with placenta previa. Magn Reson Imaging. 2019;64:71\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Hu M, Wen X, Huang Y, Luo R, Chen J. MRI-based radiomics nomogram in patients with high-risk placenta accreta spectrum: can it aid in the prenatal diagnosis of intraoperative blood loss? \u003cem\u003eAbdominal radiology (New York)\u003c/em\u003e 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong Z, Wang P, Zou L, Zhou Y, Wang X, Liu T, Zhang D. Enhancing postpartum hemorrhage prediction in pernicious placenta previa: a comparative study of magnetic resonance imaging and ultrasound nomogram. Front Physiol. 2023;14:1177795.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJha P, Pōder L, Bourgioti C, Bharwani N, Lewis S, Kamath A, Nougaret S, Soyer P, Weston M, Castillo RP, et al. Society of Abdominal Radiology (SAR) and European Society of Urogenital Radiology (ESUR) joint consensus statement for MR imaging of placenta accreta spectrum disorders. Eur Radiol. 2020;30(5):2604\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing Y, Zhang C, Wu W, Pu J, Zhao X, Zhang H, Zhao L, Schoenhagen P, Liu S, Ma X. A radiomics model based on aortic computed tomography angiography: the impact on predicting the prognosis of patients with aortic intramural hematoma (IMH). Quant Imaging Med Surg. 2023;13(2):598\u0026ndash;609.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllen L, Jauniaux E, Hobson S, Papillon-Smith J, Belfort MA. FIGO consensus guidelines on placenta accreta spectrum disorders: Nonconservative surgical management. Int J Gynaecol Obstet. 2018;140(3):281\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshirsaz AA, Fox KA, Salmanian B, Diaz-Arrastia CR, Lee W, Baker BW, Ballas J, Chen Q, Van Veen TR, Javadian P, et al. Maternal morbidity in patients with morbidly adherent placenta treated with and without a standardized multidisciplinary approach. Am J Obstet Gynecol. 2015;212(2):e218211\u0026ndash;219.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilver RM, Fox KA, Barton JR, Abuhamad AZ, Simhan H, Huls CK, Belfort MA, Wright JD. Center of excellence for placenta accreta. Am J Obstet Gynecol. 2015;212(5):561\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng L, Zhang X, Liu J, Liu Y, Huang J, Chen J, Su Y, Yang Z, Song T. MRI-radiomics-clinical-based nomogram for prenatal prediction of the placenta accreta spectrum disorders. Eur Radiol. 2022;32(11):7532\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBartels HC, O'Doherty J, Wolsztynski E, Brophy DP, MacDermott R, Atallah D, Saliba S, Young C, Downey P, Donnelly J, et al. Radiomics-based prediction of FIGO grade for placenta accreta spectrum. Eur Radiol Exp. 2023;7(1):54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDo QN, Lewis MA, Xi Y, Madhuranthakam AJ, Happe SK, Dashe JS, Lenkinski RE, Khan A, Twickler DM. MRI of the Placenta Accreta Spectrum (PAS) Disorder: Radiomics Analysis Correlates With Surgical and Pathological Outcome. J Magn Reson Imaging. 2020;51(3):936\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Hu M, Wen X, Huang Y, Luo R, Chen J. MRI-based radiomics nomogram in patients with high-risk placenta accreta spectrum: can it aid in the prenatal diagnosis of intraoperative blood loss? Abdom Radiol (NY). 2023;48(3):1107\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatsunaga S, Masuko H, Takai Y, Kanayama N, Seki H. Fibrinogen may aid in the early differentiation between amniotic fluid embolism and postpartum haemorrhage: a retrospective chart review. \u003cem\u003eSpringer Science and Business Media LLC\u003c/em\u003e 2021(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMu\u0026ntilde;oz M, Stensballe J, Ducloy-Bouthors AS, Bonnet MP, De Robertis E, Fornet I, Goffinet F, Hofer S, Holzgreve W, Manrique S, et al. Patient blood management in obstetrics: prevention and treatment of postpartum haemorrhage. A NATA consensus statement. Blood Transfus. 2019;17(2):112\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu Y, Wang Y, Weng Q, Wu X, Xia S, Wang H, Cheng X, Mao C, Li X, Zhou Z, et al. Intraplacental T2-hypointense bands may help predict placental invasion depth and postpartum hemorrhage in placenta accrete spectrum disorders in high-risk gravid patients. Magn Reson Imaging. 2022;94:73\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen T, Xu XQ, Shi HB, Yang ZQ, Zhou X, Pan Y. Conventional MRI features for predicting the clinical outcome of patients with invasive placenta. Diagn Interv Radiol. 2017;23(3):173\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBourgioti C, Konstantinidou AE, Zafeiropoulou K, Antoniou A, Fotopoulos S, Theodora M, Daskalakis G, Nikolaidou ME, Tzavara C, Letsika A, et al. Intraplacental Fetal Vessel Diameter May Help Predict for Placental Invasiveness in Pregnant Women at High Risk for Placenta Accreta Spectrum Disorders. Radiology. 2021;298(2):403\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel-Lippmann KK, Planz VB, Phillips CH, Ohlendorf JM, Zuckerwise LC, Moshiri M. Placenta Accreta Spectrum Disorders: Update and Pictorial Review of the SAR-ESUR Joint Consensus Statement for MRI. Radiographics. 2023;43(5):e220090.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Patient clinical characteristics in the training and validation cohorts.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"740\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.27027027027027%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTraining Cohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003en\u003c/strong\u003e\u003cstrong\u003e＝\u003c/strong\u003e\u003cstrong\u003e105\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.945945945945947%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIndependent validation\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ecohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003en=28\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.945945945945947%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eExternal validation cohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003en=41\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"13.2569558101473%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH(\u003c/strong\u003e\u003cstrong\u003e＋\u003c/strong\u003e\u003cstrong\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en=38\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.292962356792144%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e-\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en=67\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.474631751227497%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.2569558101473%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH(\u003c/strong\u003e\u003cstrong\u003e＋\u003c/strong\u003e\u003cstrong\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en=12\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.111292962356792%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e-\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en\u003c/strong\u003e\u003cstrong\u003e＝\u003c/strong\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.364975450081833%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.292962356792144%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH(\u003c/strong\u003e\u003cstrong\u003e＋\u003c/strong\u003e\u003cstrong\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en=20\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.075286415711947%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPH\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e-\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003en=21\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.873977086743044%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMaternal age, years\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Mean\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003eSD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(32.00-38.00)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e32.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(29-38)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.108\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e32.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(28.5-35.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(29.25-36.75)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.701\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(29.25-38)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.00\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e28.5-37\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.647\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGestational age at MRI, weeks\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Mean \u0026plusmn;SD )\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(33-36)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(32-36)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.985\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(32-34.75)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e34.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(32.23-36.00)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.673\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e32-33.75\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e32-35\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.261\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGestational age at\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eCD\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e, weeks\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Mean\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003eSD)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e36.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(35-37\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e37.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(36-37)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.037*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e36\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(35.00-37.00)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e35.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(34.25-36.75)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.433\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e36.00\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e34.25-36\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e36.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e35-37\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.425\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrevious CD history\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eNegative\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePositive\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e26\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e37\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e30\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.020*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.000\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e12\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.890\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of Previous CD\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Median,range)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(0-1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0(0-1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(0.25-1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(0-1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.688\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e0-1.75\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e0-1\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.511\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbortion history\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eNegative\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePositive\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e28\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e49\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.951\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.560\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e15\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.134\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of Previous Abortion\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Median,range)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(0-2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(0-2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.995\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1(1-2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.5(1-3)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.159\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e1-2\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e0-3\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.925\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSubtype of\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eplacenta previa\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eLow‑lying\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eMarginal\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePartial\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eComplete\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e35\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e11\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e49\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.091\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.218\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.000\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDepth of PAS invasiveness (intraoperative and/or histologic)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAccreta\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIncreta\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePercreta\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e8\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e28\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e23\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.387\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e9\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e18\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.026\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBlood transfusion\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eNegative\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ePositive\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e28\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e64\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e10\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e13\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.002*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e16\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e19\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.000*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative hemoglobin level (g/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e109.66\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e11.88\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e111.91\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e9.81\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.298\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e102.17\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e20.25\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e110.88\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e11.88\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.165\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e111.45\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e12.23\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e114.67\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e14.31\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.445\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePostoperative hemoglobin levels(g/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.53\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e14.19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e101.42\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e12.67\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e92.83\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e24.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e98.63\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e16.95\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.460\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e103.85\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e13.58\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e109.52\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e15.49\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.221\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative platelet count (\u0026times;109)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e177.63\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e54.27\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e204.33\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e64.30\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.033*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e217.00\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e50.58\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e184.25\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e35.19\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.053\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e229.71\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e76.86\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e211.40\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e52.83\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.382\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePostoperative platelet count (\u0026times;109)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e149.73\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e60.59\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e202.91\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e67.46\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e181.75\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e54.42\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e195.75\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e51.59\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.494\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e169.90\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e40.55\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e228.19\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e71.02\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.003\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative PT\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e10.97\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.51\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e10.80\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.82\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.372\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e10.81\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.92\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e10.72\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.79\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.784\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e12.87\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.37\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e12.50\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.58\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.263\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative APTT\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e27.11\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e3.20\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e26.97\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e2.84\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.820\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e26.75\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e2.91\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e26.30\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.86\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.622\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.96\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e4.04\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.51\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e3.85\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.716\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative level of\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eD-Dimer\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e5.59\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e17.05\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.90\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.26\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.082\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.54\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e2.70\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.93\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.64\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.061\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.93\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e6.05\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.16\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e4.30\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.703\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"17.43243243243243%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreoperative level of fibrinogen\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.170\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.008\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.109\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.902\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.64864864864865%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.752\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.945945945945946%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e3.79\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.83\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.10\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e0.92\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.081081081081081%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.371\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.324324324324325%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.28\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.18\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.621621621621621%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.52\u003c/strong\u003e\u003cstrong\u003e\u0026plusmn;\u003c/strong\u003e\u003cstrong\u003e1.01\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.675675675675675%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.478\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eCD cesarean delivery, SD standard deviation, PAS placenta accrete spectrum,PT prothrombin time ,APTT activated partial thromboplastin time,INR international normalized ratio, D-Dime\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Predictive performance of MRI morphological features for PPH in training, validation, and external validation cohorts\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"780\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.911424903722722%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCohort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.944801026957638%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMRI morphological findings\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.75609756097561%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.296534017971759%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensitivity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.911424903722722%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSpecificity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.810012836970476%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAccuracy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.0397946084724%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.783055198973042%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNPV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.546854942233633%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.911424903722722%\" rowspan=\"7\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTraining cohort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.944801026957638%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIntraplacental dark T2 bands\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.75609756097561%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.689\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.592-0.776)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.296534017971759%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.18\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(51.5\u0026nbsp;-\u0026nbsp;75.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.911424903722722%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.68\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(56.9\u0026nbsp;-\u0026nbsp;86.6)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.810012836970476%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e67.619\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(57.789-76.426)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.0397946084724%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e81.132\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(71.040-88.287)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.783055198973042%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e53.846\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(44.560-62.873)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.546854942233633%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePlacenta bulge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.645\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.545- 0.736)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(94.60-100.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e28.95\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(15.4\u0026nbsp;-\u0026nbsp;45.9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e74.038\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(64.519-82.143)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e70.968\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(66.616-74.965)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(71.509-100)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eLoss of retroplacental T2-hypointense line\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.732\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.636 - 0.813)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e91.045\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(81.520-96.642)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e55.263\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(38.299-71.376)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e78.095\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(68.967-85.579)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e78.205\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(71.430-83.739)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e77.778\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(60.765-88.776)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e＜\u003c/strong\u003e\u003cstrong\u003e0.0001*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMyometrial thinning/disruption\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.581\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.481 -0.677)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e92.537\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(83.437-97.533)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e23.684\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(11.444-40.241)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e67.619\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(57.789-76.426)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e68.132\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(63.879-72.103)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.286\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(39.403-83.285)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0352*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder tenting sign\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.539\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.439-0.637)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(94.643-100)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e7.895\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(1.659-21.377)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.667\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(56.799-75.567)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e65.686\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(63.559-67.753)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(29.240-46.145)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.503\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncreased placental vascularity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.711\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.614 -0.795)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e89.552\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(79.651-95.695)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e52.632\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(35.818-69.019)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e76.190\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(66.890-83.961)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e76.923\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(70.245-82.476)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e74.074\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(57.117-85.972)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.000*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbnormal vascularization of the placental bed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.561\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.460-0.658)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e91.045\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(81.520-96.642)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e23.684\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(11.444-40.241)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.667\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(56.799-75.567)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e67.778\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(63.441-71.82)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e60.000\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(36.638-79.555)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.1439\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.911424903722722%\" rowspan=\"7\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eIndependent validation\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003ecohort\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.944801026957638%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIntraplacental dark T2 bands\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.75609756097561%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.719\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.518 - 0.871)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.296534017971759%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e75.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(42.8\u0026nbsp;-\u0026nbsp;94.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.911424903722722%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e68.75\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(41.3\u0026nbsp;-\u0026nbsp;89.0)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.810012836970476%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e71.429\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(51.333-86.776)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.0397946084724%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.3\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(44.8\u0026nbsp;-\u0026nbsp;80.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.783055198973042%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e78.6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(56.6\u0026nbsp;-\u0026nbsp;91.2)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.546854942233633%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0135*\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePlacenta bulge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.594\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.393-0.774)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e25.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(5.5\u0026nbsp;-\u0026nbsp;57.2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e93.75\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(69.8\u0026nbsp;-\u0026nbsp;99.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.286\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(44.065-81.359)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e75.0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(26.2-96.2)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e62.5\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(54.0\u0026nbsp;-\u0026nbsp;70.3)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.1952\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eLoss of retroplacental T2-hypointense line\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.646\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.444 - 0.816)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e41.67\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(15.2\u0026nbsp;-\u0026nbsp;72.3)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(61.7\u0026nbsp;-\u0026nbsp;98.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e67.857\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(47.648-84.122)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e71.4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(36.8\u0026nbsp;-\u0026nbsp;91.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(54.5\u0026nbsp;-\u0026nbsp;77.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0889\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMyometrial thinning/disruption\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.646\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.444 - 0.816)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e41.67\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(15.2\u0026nbsp;-\u0026nbsp;72.3)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(61.7\u0026nbsp;-\u0026nbsp;98.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e67.857\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(47.648-84.122)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e71.4\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(36.8\u0026nbsp;-\u0026nbsp;91.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(54.5\u0026nbsp;-\u0026nbsp;77.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0889\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder tenting sign\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.583\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.383-0.765)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e8.33\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.2\u0026nbsp;-\u0026nbsp;38.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(79.4\u0026nbsp;-\u0026nbsp;100.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e42.857\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(24.462-62.821)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e90.8-100\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e42.9\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(42.9\u0026nbsp;-\u0026nbsp;42.9)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.3173\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncreased placental vascularity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.604\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.403-0.783\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.33\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(9.9\u0026nbsp;-\u0026nbsp;65.1)\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(61.7-98.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.286\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(44.065-81.359)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(30.4-90.2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e63.6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(53.0-73.1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.2090\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbnormal vascularization of the placental bed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.604\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.403 - 0.783)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e33.33\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(9.900-\u0026nbsp;65.100)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e87.50\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(61.700-98.400)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e64.286\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(44.065-81.359)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(30.40\u0026nbsp;-\u0026nbsp;90.20)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e63.6\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(53.00\u0026nbsp;-\u0026nbsp;73.10)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.2090\u003c/strong\u003e\u003c/p\u003e\u0026nbsp;\u0026nbsp;\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.911424903722722%\" rowspan=\"7\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eExternal validation cohort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.944801026957638%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIntraplacental dark T2 bands\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.75609756097561%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.777\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.620 -0.892)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.296534017971759%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e65.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(40.8\u0026nbsp;-\u0026nbsp;84.6)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.911424903722722%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e90.48\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(69.6\u0026nbsp;-\u0026nbsp;98.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.810012836970476%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e63.415\u003c/strong\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e50.895-91.343\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.0397946084724%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e86.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(62.6\u0026nbsp;-\u0026nbsp;96.2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.783055198973042%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(59.5\u0026nbsp;-\u0026nbsp;83.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.546854942233633%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.0001*\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePlacenta bulge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.604\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.439 - 0.753)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e35.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(15.4\u0026nbsp;-\u0026nbsp;59.2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e85.71\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(63.7\u0026nbsp;-\u0026nbsp;97.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e60.976\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e44.505-75.799\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e70.0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(41.1\u0026nbsp;-\u0026nbsp;88.6)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e58.1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(49.0\u0026nbsp;-\u0026nbsp;66.6)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.1236\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eLoss of retroplacental T2-hypointense line\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.631\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.466 - 0.776)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e50.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(27.2\u0026nbsp;-\u0026nbsp;72.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e76.19\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(52.8\u0026nbsp;-\u0026nbsp;91.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e63.415\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e46.936-77.877\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e66.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(45.3\u0026nbsp;-\u0026nbsp;82.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e61.5\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(49.3\u0026nbsp;-\u0026nbsp;72.5)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0790\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMyometrial thinning/disruption\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.702\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.539- 0.835)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e50.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(27.196\u0026nbsp;-\u0026nbsp;72.804)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e90.476\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(69.623\u0026nbsp;-\u0026nbsp;98.825)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e70.732\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e54.463-83.870\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e83.333\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(55.483\u0026nbsp;-\u0026nbsp;95.251)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e65.517\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(54.541\u0026nbsp;-\u0026nbsp;75.05)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0022*\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBladder tenting sign\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.524\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.362 - 0.682)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(83.2\u0026nbsp;-\u0026nbsp;100.0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e4.76\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.1\u0026nbsp;-\u0026nbsp;23.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e48.780\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e32.878-64.866\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e50.0\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(47.6\u0026nbsp;-\u0026nbsp;52.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e100\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(90.8-100)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.3173\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eIncreased placental vascularity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.777\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.620 - 0.892)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e65.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(40.8\u0026nbsp;-\u0026nbsp;84.6)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e90.48\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(69.6\u0026nbsp;-\u0026nbsp;98.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e78.049\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e62.386-89.439\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e86.7\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(62.6\u0026nbsp;-\u0026nbsp;96.2)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e73.1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(59.5\u0026nbsp;-\u0026nbsp;83.4)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.0001*\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"19.020172910662826%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbnormal vascularization of the placental bed\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.951008645533141%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.602\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(0.438 - 0.752)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.680115273775217%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e30.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(11.9\u0026nbsp;-\u0026nbsp;54.3)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.247838616714697%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e90.48\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(69.6\u0026nbsp;-\u0026nbsp;98.8)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.256484149855908%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e60.98\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e（\u003c/strong\u003e\u003cstrong\u003e44.505-75.799\u003c/strong\u003e\u003cstrong\u003e）\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.39193083573487%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e75.00\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(40.6\u0026nbsp;-\u0026nbsp;92.9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.103746397694524%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e57.60\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(49.7\u0026nbsp;-\u0026nbsp;65.1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.348703170028818%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.0985\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eModels comparison\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"714\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"21.708683473389357%\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.96358543417367%\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.742296918767507%\"\u003e\n \u003cp\u003e\u003cstrong\u003eauc\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.184873949579831%\"\u003e\n \u003cp\u003e\u003cstrong\u003eauc 95ci\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.364145658263306%\"\u003e\n \u003cp\u003e\u003cstrong\u003especifictiy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.943977591036415%\"\u003e\n \u003cp\u003e\u003cstrong\u003erecall\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.38375350140056%\"\u003e\n \u003cp\u003e\u003cstrong\u003eprecision\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.084033613445378%\"\u003e\n \u003cp\u003e\u003cstrong\u003ef-score\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.624649859943977%\"\u003e\n \u003cp\u003e\u003cstrong\u003eaccuracy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"21.708683473389357%\" rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003etraining corhort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.96358543417367%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRadiomic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.742296918767507%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.929\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.184873949579831%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.87, 0.987)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.364145658263306%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.970\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.943977591036415%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.692\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.38375350140056%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.931\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.084033613445378%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.794\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.624649859943977%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.867\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003eclinic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.734\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.631, 0.837)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.970\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.513\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.909\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.656\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.800\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003ecombine\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.956\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.91, 1.000)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.985\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.846\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.971\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.904\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.933\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"21.708683473389357%\" rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eindependence validation corhort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.96358543417367%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRadiomic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.742296918767507%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.735\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.184873949579831%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.547, 0.922)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.364145658263306%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.857\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.943977591036415%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.357\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.38375350140056%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.714\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.084033613445378%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.476\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.624649859943977%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.607\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003eclinic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.584\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.37, 0.798)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.929\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.286\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.800\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.421\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.607\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003ecombine\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.781\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.606, 0.955)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.857\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.500\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.778\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.609\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.679\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"21.708683473389357%\" rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eexternal validation corhort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"8.96358543417367%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRadiomic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"5.742296918767507%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.686\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.184873949579831%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.522, 0.850)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.364145658263306%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.810\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.943977591036415%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.400\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"9.38375350140056%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.665\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"10.084033613445378%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.500\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.624649859943977%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.610\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003eclinic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.694\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.531, 0.857)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.952\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.450\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.900\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.600\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.707\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"11.449016100178891%\"\u003e\n \u003cp\u003e\u003cstrong\u003ecombine\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"7.334525939177102%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.702\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"15.56350626118068%\"\u003e\n \u003cp\u003e\u003cstrong\u003e(0.541, 0.864)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.237924865831843%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.571\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.701252236135957%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.700\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.985688729874777%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.609\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.880143112701253%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.651\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"14.847942754919499%\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.634\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Magnetic resonance imaging, Placenta accreta spectrum, Nomogram, Postpartum hemorrhage","lastPublishedDoi":"10.21203/rs.3.rs-4550980/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4550980/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e \u003cp\u003eTo develop and validate the nomogram by combining MRI-derived radiomics and clinical features for predicting postpartum hemorrhage in high-risk placenta accreta spectrum (PAS) patients before cesarean section.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThe T2WI sagittal MR images and clinical data of 70 postpartum hemorrhage (+) pregnant women and 104 postpartum hemorrhage (-) pregnant women were retrospectively collected from two centers. These pregnancies were divided into a training (n\u0026thinsp;=\u0026thinsp;105), an independent validation (n\u0026thinsp;=\u0026thinsp;28), and an external validation (n\u0026thinsp;=\u0026thinsp;41) cohort. Radiomic features were extracted, and radiomics signature were constructed. Clinical features were analyzed retrospectively. The clinical model, the radiomic model, and the clinicoradiomic model were compared. The nomogram of the optimal model was constructed to predict the risk of postpartum hemorrhage.The diagnostic performance was assessed using the area under the receiver operating characteristic curve (AUC) and the DeLong test. Decision curve analysis (DCA) was performed to determine the performance of the best predictive model.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eEighteen radiomic features showed a strong correlation with PPH. Four MRI features were selected as clinical features.The clinicoradiomic model resulted in the best discrimination ability for risk prediction of PPH, with an AUC of 0.956 (95% CI, 0.9101.000), 0.781(95% CI, 0.606 0.955), and 0.702 (95% CI, 0.541 0.864) in the training, independent validation and external validation cohorts respectively. The clinicoradiomic nomogram, incorporating radiomics signature and four MRI features, was developed. The calibration was good and DCA confirmed the clinical utility of the nomogram.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eObstetricians can use the nomogram to noninvasively predict PPH and guide them in creating reasonable preoperative treatment plans.\u003c/p\u003e","manuscriptTitle":"MRI–radiomics–clinical–based nomogram for prediction postpartum hemorrhage in patients with suspected placenta accrete spectrum before cesarean section","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-16 21:58:29","doi":"10.21203/rs.3.rs-4550980/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"18607d7a-3ab0-4755-8667-348daec4f6ac","owner":[],"postedDate":"July 16th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-07-16T21:58:32+00:00","versionOfRecord":[],"versionCreatedAt":"2024-07-16 21:58:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4550980","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4550980","identity":"rs-4550980","version":["v1"]},"buildId":"oE6Zbj460LM0Up2FdVbMZ","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}