Development and validation of CT-based clinical-radiomics nomogram for predicting abdominal aortic aneurysms progression

Development and validation of CT-based clinical-radiomics nomogram for predicting abdominal aortic aneurysms progression | 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 Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development and validation of CT-based clinical-radiomics nomogram for predicting abdominal aortic aneurysms progression Ru Tan, Maobo Wang, Bing Kang, Xinxin Yu, Guohua Zhao, Shuai Zhang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6439190/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Background: The rapid progression associated with abdominal aortic aneurysms (AAA) is the major contributor to morbidity and mortality among the elderly. More comprehensive radiomics-based prediction of AAA progression is both valuable and warranted. Methods: In this multicenter retrospective investigation, 166 AAA patients who received two contrast-enhanced abdominal CT examinations from January 2014 to July 2022 were divided into training (n = 92) and external test cohort (n = 74). A clinical model for predicting AAA progression was built using clinical and CT characteristics that were significant independent predictors. Radiomics features were extracted from CT images, and a radiomics signature was constructed. The nomogram model was constructed by combining clinical model and radiomics signature. The diagnostic performance were evaluated and validated on the training and test sets, and then compared among the three models. Results: Over a median of 0.94 years (range, 0.5-7.05 years), 66 patients (training set: n = 35; external test set: n= 31) experienced AAA progression. The clinical model was built by diabetes and AAA maximal diameter. Seven features were employed to build the radiomics signature. In the external test set, the area under the curve was higher for the nomogram (0.83) than for the clinical model (0.69, p =0.02), and the nomogram model showed a sensitivity, specificity, and accuracy of 74.2%, 81.4%, and 78.4%. Conclusions: The nomogram model combining the clinical factors and radiomics signature showed better diagnostic performance than the clinical model, and may assist clinical decision-making process. Abdominal aortic aneurysm Radiomics Aneurysm progression computer tomography angiography nomogram Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Abdominal aortic aneurysms (AAA) are a widespread arterial dilative and degenerative disease 1 . The rapid progression and rupture associated with AAA is the major contributor to morbidity and mortality among the elderly, and its mortality rate is as high as 80% 2, 3 . Hence, evaluation of AAA progression and rupture risk is critical to AAA management. Several investigations demonstrated a strong association between AAA diameter and AAA progression and rupture. Moreover, AAA diameter has long been used in the clinics as a robust indicator of AAA progression and rupture 4 , 5 . Current guidelines indicate that AAAs > 5.5 cm in diameter require surgical intervention, whereas AAAs < 5.5 cm are closely monitored 6 . However, it is not uncommon for some smaller aneurysms to progress rapidly and rupture, thus, illustrating the limitations of using vessel diameter alone to assess AAA progression and rupture. However, in case of smaller aneurysms, effective follow-up and prediction are more meaningful. Therefore, there is a growing research interest in potential markers that can predict rapid AAA progression and rupture, such as, intraluminal thrombus (ILT), vessel wall stress, circulating markers, and so on 5 , 7 – 9 . Artificial intelligence holds great potential in transforming health care and medical imaging, and radiomics analysis is a commonly employed artificial intelligence tool for meaningful interpretation of medical data 10 . Using radiomics analysis, one can retrieve quantitative features from traditional medical images, thus overcoming limitations in visual image processing 11 , 12 . In AAA radiology, radiomics analysis is typically used to estimate AAA endoleaks and progression following endovascular aneurysm repair 13 , 14 . Hirata et al. and Lee et al. employed this tool to predict AAA expansion 15 , 16 . However, more comprehensive radiomics-based prediction of AAA progression is both valuable and warranted. Herein, the goal was to establish and verify a CT-based clinical-radiomics nomogram for the prediction of AAA progression. 2. Materials and Methods 2.1 Patients This retrospective work received institutional approval, and the informed consent requirement was waived. In detail, ethical approval was obtained from the Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong First Medical University, China. In addition, patient informed consent was waived due to the retrospective nature of this study, which was under the permission of the Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong First Medical University. All methods were carried out in accordance with relevant guidelines and regulations. Clinical trial number: not applicable. For this investigation, we recruited AAA patients via the medical/radiological reports of three separate hospitals between January 2014 and July 2022. The following patients were included in our analysis: (1) AAA patients who underwent two contrast-enhanced abdominal CT examinations with a time interval > 6 months; and (2) no endovascular or open aortic repair before or during the aforementioned imaging interval. The following patients were excluded from analysis: (1) patients suffering from other aortic diseases that impacted image analysis, such as, aortic dissection, aortic occlusion, and so on; (2) missing clinical information; and (3) poor image quality. Patient clinical information was retrieved from the hospital medical records, including patient age, sex, smoking history, hypertension, diabetes, hyperlipidemia, anti-hypertensive treatment, and lipid-lowering treatment. In all, 166 patients (147 men; average age, 69.2 ± 10.2 years; 66 with AAA progression and 100 without AAA progression) were recruited from analysis. Among them, 92 patients (35 with AAA progression and 57 without AAA progression) from one hospital (Shandong Provincial Hospital) served as the training cohort. Additionally, 74 patients (31 with AAA progression and 43 without AAA progression) from the two remaining hospitals (Qilu Hospital of Shandong University; Shandong Provincial Qianfoshan Hospital) formed the external test cohort. 2.2 CT imaging protocol All contrast-enhanced abdominal CT images were acquired from the multidetector CT scanners (Somatom Force, Siemens Healthcare; Somatom Definition Flash, Siemens Healthcare; Ingenuity CT, Philips). Enhanced CT scanning encompassed the base of the neck to the aortoiliac bifurcation, with the patient positioned supine. Next, using a power injector, 90–120 mL of contrast media (Omnipaque-350; GE Healthcare) was administered at 4-4.5 mL/s, before flushing with 40 mL saline at the same speed. Subsequently, bolus tracking was employed to acquire data at an attenuation cut-off of 100 Hounsfield units at the celiac trunk level for 6 s. Scanning was proceeded as follows: tube voltage: 120 kVp, pitch: 0.8-1.0, tube current: 300–400 mA (via automatic tube current modulation), matrix: 512 × 512, recreated slice thickness: 1 mm, recreated slice interval: 1 mm, rotation duration: 500–600. 2.3 CT imaging analysis and clinical model progression The baseline CT imaging features were retrieved, including, maximum AAA diameter and total volume; with or without ILT; as well as maximum ILT diameter and volume. In addition, we also obtained the total AAA volume at the follow-up CT. ILT presence was described as ILT thickness, along with a > 5 mm aortic wall on enhanced CT. 7 Maximum AAA and ILT diameters were assessed based on the maximum cross-sections or maximum ILT cross-sections perpendicular to the long axis of the aorta using a multiplanar reconstruction method. The superior and inferior AAA borders were defined as the loss of aortic wall parallelism to the aortic dilatation end. The AAA progression rate was computed using the following formula: (AAA volume at follow-up - AAA volume at baseline) / follow-up years. AAA progression was defined as an annual AAA progression rate ≥ 10 mL/year. 8 All imaging features were assessed and quantified by two cardiovascular radiologists (Z.S. and H.G., with 7 and 10 years of experience in vascular imaging, respectively). The evaluators were blinded to the subjects’ clinical data using semiautomatic segmentation via the ITK-SNAP software (version 3.8.0, open source, http://www.itksnap.org ). We performed inter-patient comparisons of the clinical and CT features among participants with and without AAA progression using univariable analysis (quantitative data was assessed via independent-sample t tests, whereas, qualitative data via chi-square tests). Variables reaching significance in univariable analysis were entered into in multivariable logistic regression analysis to estimate AAA progression. Lastly, the significant independent AAA indicators from multivariable analysis were employed for clinical model construction. 2.4 Lesion segmentation and radiomics feature retrieval A flowchart depicting our radiomics analysis is presented in Fig. 1 . A post-processing platform (Lianying Intelligent Medical Technology Co., Ltd., Beijing, China) was employed for lesion segmentation and radiomics feature acquisition. Prior to radiomics feature extraction, images were batch adjusted, namely, gray-scale discretization and image resampled, to manage differences in image quality and noise, and reduce radiomics features variability. Using manual segmentation, we next selected regions of interests (ROIs) in the lesion cross-sectional area. Contours were drawn within 5 mm of the aortic wall avoiding other organ tissues. Lastly, we extracted several radiomics variables, such as, the first-order statistics, shape, and textural features (gray-level size zone, gray-level co-occurrence, gray-level dependence, gray-level run-length, and neighborhood gray-tone differencematrix features). To assess the inter- and intra-observer reproducibility of the aforementioned feature acquisitions, we employed the inter- and intra-class correlation coefficients (ICCs). 30 CT images (17 AAA progression and 13 AAA non-progression) were arbitrarily chosen for ROI segmentation by two independent radiologists (S.Z. and M.H.L.). Moreover, M.H.L. repeated the segmentation 3 weeks after the first acquisition to evaluate reproducibility of the extraction. An ICC > 0.75 represented satisfactory agreement of feature acquisition. The remaining image segmentations were subsequently conducted by M.H.L, with proper blinding to patient clinical information. 2.5 Feature selection and radiomics signature and nomogram model progression We first selected radiomics features with ICCs > 0.75. Then, the remaining features were tested via the select_k_best method to identify potential significant features. Finally, the least absolute shrinkage and selection operator (LASSO) regression model was employed for the selection of the most significant features. Features identified in LASSO were utilized in radiomics signature model construction, and an radiomics score (Rad-score) was computed for individuals using a linear combination of identified features weighted by their corresponding LASSO coefficients. A radiomics-based nomogram was then established by incorporating the most relevant variables from the clinical model and Rad-score. 2.6 Statistical Analysis Receiver operating characteristic curve (ROC) analysis was employed for the measurement of the areas under the curve (AUCs) for predicting AAA progression in the training sets and external test sets of the clinical model, radiomics signature, and nomogram model. AUC comparisons were conducted via the Delong test. The predictive power of the nomogram model was assessed via the decision curve analysis (DCA). nomogram model calibration was performed using calibration curves. Two-sided p < 0.05 was deemed as significant. SPSS (version 24.0, IBM) and R (version 3.3.3, https://www.r-project.org ) were employed for all data analyses. 3. Results 3.1 clinical model In all, 166 individuals (147 men; average age, 69.2 ± 10.2 years) were included in our analysis. The median follow-up duration was 0.94 years (ranging between 0.5–7.05 years) with a progression rate of 16.9 ± 28.8 mL/y. Among them, 92 patients (35 with AAA progression and 57 without AAA progression) served as the training set, and 74 patients (31 with AAA progression and 43 without AAA progression) as the external test set. Clinical and CT features of patients from both groups in the training set and external test set are presented in Table 1 . In the training set, AAA progression patients, relative to AAA non-progression patients, revealed enhanced diabetes incidences (42.9% vs. 12.3%, p = 0.001), larger AAA maximal diameter (44.2 ± 5.3 mm vs. 39.2 ± 6.4 mm, p 0.05). Using multivariate analysis, we revealed that diabetes [odds ratio (OR): 3.77; 95% confidence interval (CI): 1.25–11.42; p = 0.02] and AAA maximal diameter (OR: 1.11; 95% CI: 1.02–1.20; p = 0.02) were the independent predictors of AAA progression. Hence, these variables were employed for clinical model construction. Table 1 Clinical characteristics and CT imaging characteristics of patients with and without AAA progress Parameter Training Set (n = 92) External test Set (n = 74) Non-progression Group (n = 57) Progression Group (n = 35) P Non-progression Group (n = 43) Progression Group (n = 31) P Age (y) 69.3 ± 9.6 68.1 ± 8.6 0.56 68.5 ± 12.3 71.6 ± 9.7 0.25 Sex, male 49 (86.0) 31 (88.6) 0.72 38 (88.4) 29 (93.5) 0.45 Hypertension 41 (71.9) 26 (74.3) 0.81 31 (72.1) 21 (67.7) 0.69 Diabetes 7 (12.3) 15 (42.9) 0.001 12 (27.9) 9 (29.0) 0.92 Hyperlipidemia 24 (42.1) 12 (34.3) 0.46 16 (37.2) 8 (25.8) 0.30 Smoking history 28 (49.1) 19 (54.3) 0.63 30 (69.8) 19 (61.3) 0.45 Anti-hypertensive treatment 44 (77.2) 28 (80.0) 0.75 33 (76.7) 23 (74.2) 0.80 Lipid-lowering treatment 26 (45.6) 13 (37.1) 0.43 19 (44.2) 10 (32.3) 0.30 Maximal diameter (mm) 39.2 ± 6.4 44.2 ± 5.3 < 0.001 39.7 ± 6.1 44.3 ± 6.8 0.003 Total aneurysm volume (mL) 51.4 ± 44.1 79.8 ± 66.9 0.03 65.2 ± 58.8 101.2 ± 77.7 0.03 Presence of ILT 42 (73.7) 31 (88.6) 0.09 31 (72.1) 22 (71.0) 0.92 ILT maximal diameter (mm) 8.5 ± 7.3 11.2 ± 7.1 0.08 11.3 ± 9.9 8.7 ± 8.9 0.03 ILT volume (mL) 22.5 ± 28.6 33.5 ± 44.6 0.16 26.2 ± 53.0 22.4 ± 26.7 0.19 Continuous variables are described as mean ± standard deviation, and categorical variables are presented as numbers (%). AAA, abdominal aortic aneurysm; ILT, Intraluminal thrombus. 3.2 Development of radiomics signature and nomogram model Overall, 2264 radiomics features were obtained from the CT images. Among them, 1412 were selected following elimination of features with ICC < 0.75. Out of the 1412, 313 revealed obvious differences between the two patient populations ( p < 0.05) as evidenced by the select_best_k method. Among them, 9 radiomics features were selected as the most relevant, using LASSO. Radiomics signature was generated based on these 9 radiomics features (termed A-I) and their matched LASSO coefficients as follows: 0.40 + 0.074 × A + 0.069 × B + 0.046 × C + 0.044 × D + E × 0.042 + F × 0.038 - G × 0.033 - H × 0.038 - I × 0.061 (Table 2 ). Out of the 9 radiomics features, three exhibited the highest coefficients, and they were shape_maximum2ddiameterslice, firstorder_wavelet-hll-kurtosis, and firstorder_wavelet-hlh-skewness. In the external test set, the shape_maximum 2d diameterslice (64.9 ± 18.1 vs. 55.2 ± 9.1; p = 0.004), firstorder_wavelet-hll-kurtosis (30.2 ± 13.5 vs. 20.5 ± 14.7; p = 0.005), and firstorder_wavelet-hlh-skewness (0.8 ± 0.3 vs. 0.6 ± 0.3; p = 0.009) were obviously enhanced among AAA progression patients, compared to the AAA non-progression patients (Table 2 ). Table 2 Radiomics features selected for inclusion in radiomics signature. Variable Feature Type Feature LASSO Coefficient A Shape maximum2ddiameterslice 0.074 B First order wavelet-hll-kurtosis 0.069 C GLSZM largearealowgraylevelemphasis 0.046 D First order Normalize mean 0.044 E GLSZM log-sigma-2-mm-3d-smallareahighgraylevelemphasis 0.042 F First order wavelet-llh-totalenergy 0.038 G GLSZM Boxsigmaimage zonepercentage -0.033 H Shape sphericity -0.038 I First order wavelet-hlh-skewness -0.061 GLSZM = gray level size zone matrix, LASSO = least absolute shrinkage and selection operator. Moreover, the radiomics signature was considerably high among AAA progression patients, compared to AAA non-progression patients within the training set (2.0 ± 0.7 vs. 1.1 ± 0.5, p < 0.001) and external test set (1.8 ± 0.9 vs. 1.0 ± 0.6, p < 0.001). The AAA progression-predicting nomogram model was generated by incorporating diabetes, AAA maximal diameter, and radiomics signature (Fig. 2 ). Based on the calibration curves, this nomogram model exhibited satisfactory calibration in the training set and external test set (Fig. 2 ). 3.3 Performance evaluation The predictive abilities of clinical model, radiomics signature, and nomogram model in diagnosing AAA progression in the training set and external test set are presented in Table 3 . Inter-AUC differences among models are also provided in Table 3 . Lastly, the ROC curves of all models in the training set and external test set are presented in Fig. 3 . Table 3 Diagnostic performance of clinical model, radiomics signature, and nomogram Measure Clinical Model [1] Radiomics Signature [2] Nomogram [3] p 1 vs 2 1 vs 3 2 vs 3 Training set AUC 0.79 0.86 0.89 0.26 0.02 0.20 Accuracy 72.8 (67/92) 77.2 (71/92) 83.7 (77/92) - - - Sensitivity 82.9 (29/35) 80.0 (28/35) 77.1 (27/35) - - - Specificity 66.7 (38/57) 75.4 (43/57) 87.7 (50/57) - - - External test set AUC 0.69 0.77 0.83 0.36 0.02 0.12 Accuracy 68.9 (51/74) 71.6 (53/74) 78.4 (58/74) - - - Sensitivity 45.2 (14/31) 83.9 (26/31) 74.2 (23/31) - - - Specificity 86.1 (37/43) 62.8 (27/43) 81.4 (35/43) - - - Note—Accuracy, sensitivity, and specificity reported as percentages with numerators and denominators in parentheses. AUC, Area under the curve. In the training set, the AUCs were 0.79, 0.86, and 0.89 for the clinical model, radiomics signature, and nomogram model, respectively. The corresponding accuracy, sensitivity, and specificity for clinical model were 72.8% (67 of 92 patients), 82.9% (29 of 35 patients), and 66.7% (38 of 57 patients), respectively; for radiomics signature, 77.2% (71 of 92 patients), 80.0% (28 of 35 patients), and 75.4% (43 of 57 patients), respectively; and for nomogram model, 83.7% (77 of 92 patients), 77.1% (27 of 35 patients), and 87.7% (50 of 57 patients), respectively. The nomogram model AUC was considerably higher, compared to the clinical model AUC ( p = 0.02). However, no obvious differences were observed in the AUCs between radiomics signature and those of clinical model and nomogram model ( p = 0.26 and p = 0.20, respectively). In the external test set, the AUCs were 0.69, 0.77, and 0.83 for the clinical model, radiomics signature, and nomogram model, respectively. The accuracy, sensitivity, and specificity for clinical model were 68.9% (51 of 74 patients), 45.2% (14 of 31 patients), and 86.1% (37 of 43 patients), respectively; for radiomics signature were 71.6% (53 of 74 patients), 83.9% (26 of 31 patients), and 62.8% (27 of 43 patients), respectively; and for nomogram model were 78.4% (58 of 74 patients), 74.2% (23 of 31 patients), and 81.4% (35 of 43 patients), respectively. The nomogram model AUC was considerably higher, compared to the clinical model AUC ( p = 0.02). However, no obvious differences were observed in the AUCs between radiomics signature and those of clinical model and nomogram model ( p = 0.36 and p = 0.12, respectively). The DCAs of all models are provided in Fig. 4 . Based on our analysis, relative to clinical model, nomogram model exhibited superior overall benefit in delineating between AAA progression and AAA non-progression patients over a wide range of cut-off possibilities at which an AAA progression diagnosis is likely to occur. Figure 5 illustrates the CT images of representative patients with and without AAA progression, and their corresponding nomogram results. 4. Discussion AAA is a widespread disease with high mortality 1 , 2 . Hence, the accurate prediction of AAA progression is vital to the clinical management and intervention of this disease 17 . Herein, we established and verified a CT-guided nomogram of AAA progression estimation by incorporating clinical model and radiomics signature data. The nomogram model demonstrated satisfactory diagnostic performance in the training set (AUC, 0.83; sensitivity, 74.2%; specificity, 81.4%; and accuracy, 78.4%), and was superior to clinical model ( P = 0.02) in the external test set. These findings suggested that the clinical-radiomics nomogram is an effective tool for predicting AAA progression in the clinics. The AAA diameter is a widely investigated image-based predictive factor that is closely associated with progression and rupture, and is the main indication for treatment 18 . Several earlier studies revealed that a larger initial diameter results in larger expansion rates and is intricately linked to AAA progression 5 , 19 . This is consistent with the conclusions of our study, whereby we demonstrated that the AAA maximal diameters (OR: 1.11; 95% CI: 1.02–1.20; P = 0.02) were independent indicators of AAA progression. Moreover, we identified that diabetes (OR: 3.774; 95% CI: 1.25–11.42; P = 0.02) was another independent indicator of AAA progression. A possible reason for this is that hyperglycemia damages the inner blood vessel wall by increasing local inflammatory factor and free radical accumulations, which, in turn, inflicts damage to the vascular endothelial structure and overall integrity of the blood vessel wall, thereby promoting AAA progression 20 . Several reports demonstrated a strong relation between ILT and AAA progression, and revealed that the ILT size and volume are strongly linked to AAA occurrence 21 , 22 . Interestingly, herein, we did not observe an obvious difference in the ILT diameter and volume between the progressive and non-progressive cohorts. The reason for this difference may be our relatively smaller sample size. Artificial intelligence analysis is an emerging noninvasive tool for the extraction of quantitative features from images, thus compensating for the limitations of visual image evaluation 11 , 12 . At present, there is growing interest in research involving the detection of disease, cardiovascular events prediction analysis, and assessment of prognosis in cardiovascular radiology 13 , 23 , 24 . Lee et al. 16 examined 94 AAA patients using machine learning to predict AAA progression. Moreover, Wang et al. 13 established models for estimating prognosis following endovascular AAA repair based on morphology, deep learning, and radiomics signature. Herein, we generated a clinical-radiomics nomogram model to predict AAA progression, and our model displayed satisfactory diagnostic performance (AUC: 0.83, in external test set). Herein, the radiomics feature with the highest LASSO coefficient was Shape_maximum2ddiameterslice, which reflected lesion morphology. This was similar to the clinical model in our study, whereby we concluded that the AAA progression patients possessed larger AAA maximal diameters than non-AAA progression patients. Thus, diameter was a major parameter predicting AAA progression. However, the machine automatically measured Shape_maximum2ddiameterslice on horizontal images, whereas, clinical model measurements were made in the cross-sections perpendicular to the long aortic axis using a multiplanar reconstruction method. Moreover, the radiomics features firstorder_wavelet-hll-kurtosis and firstorder wavelet hlh skewness were significantly enhanced among AAA progression patients, suggesting greater inhomogeneity of image pixels 25 . This difference may be attributed to the greater blood flow complexity or heterogeneity of surrounding tissue among AAA progression patients. In addition, several studies revealed that the computational fluid dynamics parameters and perivascular inflammation may also predict AAA progression 8 , 26 . Herein, we selected the CT angiography (CTA) as our assessment modality. Relative to ultrasound, CTA employs three-dimensional reconstruction, which greatly reduces subjectivity influence 27 . In addition, compared to the magnetic resonance imaging examination, CTA scanning provides faster results, and avoids MRI contraindications, such as, claustrophobia, cardiac pacemaker, and so on 27 . Given that the contrast agent may conceal the textural data of parts of images during CTA examination, we acquired > 5 mm area around the AAA. Several investigations demonstrated that 5 mm of adipose tissue surrounding the AAA reflects the degree of aneurysm-based inflammation, which eventually influences AAA progression 26 . Multiple studies employed alterations in the vessel diameter to define AAA progression 28 , 29 . Although vessel diameter is easy to measure, the volume is more sensitive to alterations than diameter 30 , 31 . Therefore, similar to other studies, we employed alterations in the vessel volume to define AAA progression. This investigation has certain limitations. First, being a retrospective study, there may be certain unintentional selection bias. Second, despite being a multicenter study, our sample size was relatively small. Thus, we recommend future investigations with a larger patient population and longer follow-up duration. Third, in an attempt to minimize radiation from plain scan, most patients only underwent CTA examination. However, given that the contrast agent may conceal the textural information of images, a plain scan would have provided a certain research value, and would be the content of our next investigation. Forth, CT features were not examined at follow-ups, and there was no assessment of alterations between pre- and post-CT images. We consider that the baseline CT features are more important in predicting AAA progression, and our established model constructed with the baseline CT feature demonstrated satisfactory diagnostic efficacy. 5. Conclusions In conclusion, herein, we demonstrated that the nomogram model, which incorporated data from the clinical model and radiomics signature, displayed superior diagnostic performance than the clinical model. Based on our conclusions, the CT-based clinical-radiomics nomogram may be a robust tool for the prediction of AAA progression, and may accurately guide clinical management in the near future. Abbreviations AAA Abdominal aortic aneurysm AUC Area under the curve CI Confidence interval CTA CT angiography DCA Decision curve analysis ICC Inter and intra-class correlation coefficients ILT Intraluminal thrombus LASSO Least absolute shrinkage and selection operator OR Odds ratio ROI Regions of interest. Declarations Funding The present study was supported by the National Natural Science Foundation of China (grant no. 81871354, 81571672), and Academic promotion programme of Shandong First Medical University (grant no. 2019QL023). Declaration of Competing Interest The authors declare no competing interests. Contributions SZ and XMW were involved in conceptualization, writing-original Draft, writing-review and editing; RT was involved in methodology, data Curation, and formal analysis; MBW and BK were involved in formal analysis and funding acquisition; XXY and GHZ involved in investigation, data curation and project administration; MBW involved in methodology and software and validation; XMW involved in resources, visualization, supervision and funding acquisition. All authors read and approved the final manuscript. Availability of data and materials The data used and analyzed during the current study are available from the corresponding author on reasonable request. Conflict of Interest Statement The authors have no conflicts of interest to declare. Consent to Publish declaration: Not applicable. Acknowledgements None. References Schanzer A, Oderich GS. Management of Abdominal Aortic Aneurysms. N Engl J Med. 2021;385:1690–8. US Preventive Services Task Force, Owens DK, Davidson KW, et al. Screening for Abdominal Aortic Aneurysm: US Preventive Services Task Force Recommendation Statement. JAMA. 2019;322:2211–8. Guirguis-Blake JM, Beil TL, Senger CA, Coppola EL. Primary Care Screening for Abdominal Aortic Aneurysm: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2019;322:2219–38. Collaborators RESCAN, Bown MJ, Sweeting MJ, Brown LC, Powell JT, Thompson SG. Surveillance intervals for small abdominal aortic aneurysms: a meta-analysis. JAMA. 2013;309:806–13. Kosmas I, Paraskevas A, Mansilha. Implications of abdominal aortic aneurysm rupture at a lower diameter than the recommended threshold for AAA repair. Int Angiol. 2023;42(4):279–81. Chaikof EL, Dalman RL, Eskandari MK, et al. The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vasc Surg. 2018;67:2–e772. Zhu C, Leach JR, Wang Y, Gasper W, Saloner D, Hope MD. Intraluminal Thrombus Predicts Rapid Growth of Abdominal Aortic Aneurysms. Radiology. 2020;294:707–13. Meyrignac O, Bal L, Zadro C, et al. Combining Volumetric and Wall Shear Stress Analysis from CT to Assess Risk of Abdominal Aortic Aneurysm Progression. Radiology. 2020;295:722–9. Fan YN, Ke X, Yi ZL, et al. Plasma D-dimer as a predictor of intraluminal thrombus burden and progression of abdominal aortic aneurysm. Life Sci. 2020;240:117069. Rauschecker AM, Rudie JD, Xie L, et al. Artificial Intelligence System Approaching Neuroradiologist-level Differential Diagnosis Accuracy at Brain MRI. Radiology. 2020;295:626–37. Kosmas I, Paraskevas; Armando M. Implications of abdominal aortic aneurysm rupture at a lower diameter than the recommended threshold for AAA repair. Int Angiol. 2023;42(4):279–81. Song Z, Tang Z, Liu H, Guo D, Cai J, Zhou Z. A clinical-radiomics nomogram may provide a personalized 90-day functional outcome assessment for spontaneous intracerebral hemorrhage. Eur Radiol. 2021;31:4949–59. Wang Y, Zhou M, Ding Y, et al. Development and Comparison of Multimodal Models for Preoperative Prediction of Outcomes After Endovascular Aneurysm Repair. Front Cardiovasc Med. 2022;9:870132. Ding N, Hao Y, Wang Z, et al. CT texture analysis predicts abdominal aortic aneurysm post-endovascular aortic aneurysm repair progression. Sci Rep. 2020;10:12268. Hirata K, Nakaura T, Nakagawa M, et al. Machine Learning to Predict the Rapid Growth of Small Abdominal Aortic Aneurysm. J Comput Assist Tomogr. 2020;44:37–42. Lee R, Jarchi D, Perera R, et al. Applied Machine Learning for the Prediction of Growth of Abdominal Aortic Aneurysm in Humans. EJVES Short Rep. 2018;39:24–8. Gandhi R, Bell M, Bailey M, Tsoumpas C. Prospect of positron emission tomography for abdominal aortic aneurysm risk stratification. J Nucl Cardiol. 2021;28:2272–82. Nana P, Spanos K, Dakis K, Brodis A, Kouvelos G. Imaging Predictive Factors of Abdominal Aortic Aneurysm Growth. J Clin Med . 2021; 10:1917. Petroula Nana K, Spanos K, Dakis, et al. Imaging Predictive Factors of Abdominal Aortic Aneurysm Growth. J Clin Med. 2021;10(9):1917–27. Belén Picatoste I, Cerro-Pardo LM, Blanco-Colio, et al. Protection of diabetes in aortic abdominal aneurysm: Are antidiabetics the real effectors? Front Cardiovasc Med. 2023;10(0):1112430–41. YingNan F, Ke X, Yi ZL, et al. Plasma D-dimer as a predictor of intraluminal thrombus burden and progression of abdominal aortic aneurysm. Life Sci. 2020;240(0):117069–89. Liangliang Zhang, Byron A, Zambrano J, Choi, et al. Intraluminal thrombus effect on the progression of abdominal aortic aneurysms by using a multistate continuous-time Markov chain model. J Int Med Res. 2020;48(11):300060520968449. Guo Y, Chen X, Lin X, et al. Non-contrast CT-based radiomic signature for screening thoracic aortic dissections: a multicenter study. Eur Radiol. 2021;31:7067–76. Hu W, Wu X, Dong D, et al. Novel radiomics features from CCTA images for the functional evaluation of significant ischaemic lesions based on the coronary fractional flow reserve score. Int J Cardiovasc Imaging. 2020;36:2039–50. Meng C, Yang W, Lan H, Ren X, Li M. Development and Application of a Vehicle-Mounted Soil Texture Detector. Sens (Basel). 2020;20:7175. Dias-Neto M, Meekel JP, van Schaik TG, et al. High Density of Periaortic Adipose Tissue in Abdominal Aortic Aneurysm. Eur J Vasc Endovasc Surg. 2018;56:663–71. Timothy P. Szczykutowicz. Computed Tomography Angiography: Principles and Advances. Radiol Clin North Am. 2024;62(3):371–83. Chiara Stassi C, Mondello G, Baldino, et al. State of the Art on the Role of Postmortem Computed Tomography Angiography and Magnetic Resonance Imaging in the Diagnosis of Cardiac Causes of Death: A Narrative Review. Tomography. 2022;8(2):961–73. Le S, Wu J, Liu H, et al. Single-cell RNA sequencing identifies interferon-inducible monocytes/macrophages as a cellular target for mitigating the progression of abdominal aortic aneurysm and rupture risk. Cardiovasc Res. 2024;120(11):1351–64. Drew J, Braet TJ, Baker L, Delbono, et al. Three-dimensional characterization of sex differences in abdominal aortic aneurysm progression via vascular deformation mapping. Sci Rep. 2024;14(1):24215. Bouwens E, Vanmaele A, Sanne E, Hoeks, et al. Circulating biomarkers of cardiovascular disease are related to aneurysm volume in abdominal aortic aneurysm. Vasc Med. 2023;28(5):433–42. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 04 May, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviews received at journal 18 May, 2025 Reviewers agreed at journal 18 May, 2025 Reviewers invited by journal 14 May, 2025 Editor invited by journal 22 Apr, 2025 Editor assigned by journal 19 Apr, 2025 Submission checks completed at journal 19 Apr, 2025 First submitted to journal 13 Apr, 2025 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 Advisory Board 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-6439190","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":458246182,"identity":"525fdfb8-3b71-42c2-85af-f71a28fd08b4","order_by":0,"name":"Ru Tan","email":"","orcid":"","institution":"Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ru","middleName":"","lastName":"Tan","suffix":""},{"id":458246183,"identity":"a8b4bf56-6a74-41c1-83a5-ebdc47a9319b","order_by":1,"name":"Maobo Wang","email":"","orcid":"","institution":"Shandong Provincial Maternal and Child Health Care Hospital Affiliated to Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Maobo","middleName":"","lastName":"Wang","suffix":""},{"id":458246184,"identity":"10ba5de0-f1b5-44b5-80cc-d385ebe98246","order_by":2,"name":"Bing Kang","email":"","orcid":"","institution":"Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Bing","middleName":"","lastName":"Kang","suffix":""},{"id":458246185,"identity":"37cf3d41-4b4e-4d66-882b-af34b976d396","order_by":3,"name":"Xinxin Yu","email":"","orcid":"","institution":"Shandong Provincial Maternal and Child Health Care Hospital Affiliated to Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Xinxin","middleName":"","lastName":"Yu","suffix":""},{"id":458246186,"identity":"a77e4561-e16b-4819-9caf-bc663664973c","order_by":4,"name":"Guohua Zhao","email":"","orcid":"","institution":"Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Guohua","middleName":"","lastName":"Zhao","suffix":""},{"id":458246187,"identity":"9912f1aa-650f-473b-9f2e-c2b9d82dd549","order_by":5,"name":"Shuai Zhang","email":"","orcid":"","institution":"Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":false,"prefix":"","firstName":"Shuai","middleName":"","lastName":"Zhang","suffix":""},{"id":458246188,"identity":"3004148d-d1d7-40e4-b29d-5ef31b24aa40","order_by":6,"name":"Ximing Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIiWNgGAWjYNACAwYeBgb2gw8+VAA5EsRr4Uk2nHEGpiWBOLvMhHnbiNBicPzs4dc8BdtkzPkXpDHzzjssxz+7ge3Bxx94tJzJS7OcYXCbx3LGw2MP5247bCxx5wC74Qw8tpgdyDEz+ADUYnDjQLrB222HExtuJLBJ8+DTcv6NmUECRIuZBO+cw/XzQVr+4NNyI8f4AdiW8w1mkrwNhxMMQFrwed/+xhszxhlgW0CBfCzdcOONxDbJnjTcWiT7c4w/8/y5bW9w/jgwKmus5eVuJB+T+GGDWwsQsEGiWwLslmYgZmzAqx4ImD+AKf4DILKOkOpRMApGwSgYgQAA8Bxba3e2fQsAAAAASUVORK5CYII=","orcid":"","institution":"Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University","correspondingAuthor":true,"prefix":"","firstName":"Ximing","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-04-13 12:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6439190/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6439190/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83125801,"identity":"7e337ed5-7ec1-4b1b-bd83-fc41eef714e2","added_by":"auto","created_at":"2025-05-20 09:39:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":797095,"visible":true,"origin":"","legend":"\u003cp\u003eThe workflow of the radiomics analysis.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/9ca40888cdf8f2f740058e1f.png"},{"id":83126641,"identity":"961f8715-6aae-4aad-8af4-92c9c1a6a4b0","added_by":"auto","created_at":"2025-05-20 09:47:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":600884,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram and Calibration curves in training set.\u003c/p\u003e\n\u003cp\u003e(a) Nomogram developed in training set, which assigns points for diabetes, abdominal aortic aneurysm maximal diameter, and the radiomics signature. (b) Calibration curves for nomogram in training set.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/7ed6c07b53017b62a6ad43cc.png"},{"id":83126647,"identity":"51353c22-82a9-49eb-b1e5-d587cd8b2cd4","added_by":"auto","created_at":"2025-05-20 09:47:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1095731,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves of clinical model, radiomics signature, and nomogram model for differentiating abdominal aortic aneurysm progression in (a) training and (b) external test sets, respectively.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/5040394e5031ffc47f705948.png"},{"id":83125806,"identity":"38f85699-1575-4b5c-b024-e1acb765c208","added_by":"auto","created_at":"2025-05-20 09:39:58","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1805069,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis for three models. Y-axis indicates net benefit; x-axis indicates threshold probability. Yellow line, blue line, and green line represent net benefit of clinical model, radiomics signature, and nomogram model, respectively. Nomogram model show higher net benefit in differentiating AAA progression than clinical model across range of threshold probabilities at which a patient would be diagnosed as having AAA progression.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/ac398825cb437ac90516fc59.png"},{"id":83125805,"identity":"da5651f7-6ed2-4e09-b364-93a76fdebd08","added_by":"auto","created_at":"2025-05-20 09:39:58","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1116198,"visible":true,"origin":"","legend":"\u003cp\u003eApplication of nomogram to predict probability of AAA progression. Nomograms show points assigned for each predictor. Total points are calculated by adding points assigned for all variables; total points are then used to determine corresponding risks of AAA progression. (a, b) Coronal contrast-enhanced CT images showed that the aneurysm grew from 20.1 mL to 47.2 mL over a 2.8-year period at a growth rate of 9.8 mL/y. (c, d) Coronal contrast-enhanced CT images showed that the aneurysm grew from 44.8 mL to 101.6 mL over a 4.4-year period at a growth rate of 12.7 mL/y. (e) For patient with AAA progression (red solid arrows), nomogram yields total of 88.6 points and corresponding risk of AAA progression greater than 0.9. For patient without AAA progression (blue dashed arrows), nomogram yields total of 36.4 points and corresponding risk of AAA progression less than 0.1.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/761b954c8634fae80c1053ec.png"},{"id":83128591,"identity":"029b7641-5420-439b-bb56-aace5014ab00","added_by":"auto","created_at":"2025-05-20 09:56:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5819858,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6439190/v1/e87c212b-99da-459e-b20c-2d9f4f257d53.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and validation of CT-based clinical-radiomics nomogram for predicting abdominal aortic aneurysms progression","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAbdominal aortic aneurysms (AAA) are a widespread arterial dilative and degenerative disease\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. The rapid progression and rupture associated with AAA is the major contributor to morbidity and mortality among the elderly, and its mortality rate is as high as 80%\u003csup\u003e2, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Hence, evaluation of AAA progression and rupture risk is critical to AAA management. Several investigations demonstrated a strong association between AAA diameter and AAA progression and rupture. Moreover, AAA diameter has long been used in the clinics as a robust indicator of AAA progression and rupture\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Current guidelines indicate that AAAs\u0026thinsp;\u0026gt;\u0026thinsp;5.5 cm in diameter require surgical intervention, whereas AAAs\u0026thinsp;\u0026lt;\u0026thinsp;5.5 cm are closely monitored\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. However, it is not uncommon for some smaller aneurysms to progress rapidly and rupture, thus, illustrating the limitations of using vessel diameter alone to assess AAA progression and rupture. However, in case of smaller aneurysms, effective follow-up and prediction are more meaningful. Therefore, there is a growing research interest in potential markers that can predict rapid AAA progression and rupture, such as, intraluminal thrombus (ILT), vessel wall stress, circulating markers, and so on\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eArtificial intelligence holds great potential in transforming health care and medical imaging, and radiomics analysis is a commonly employed artificial intelligence tool for meaningful interpretation of medical data\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. Using radiomics analysis, one can retrieve quantitative features from traditional medical images, thus overcoming limitations in visual image processing\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. In AAA radiology, radiomics analysis is typically used to estimate AAA endoleaks and progression following endovascular aneurysm repair\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Hirata et al. and Lee et al. employed this tool to predict AAA expansion\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. However, more comprehensive radiomics-based prediction of AAA progression is both valuable and warranted.\u003c/p\u003e \u003cp\u003eHerein, the goal was to establish and verify a CT-based clinical-radiomics nomogram for the prediction of AAA progression.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Patients\u003c/h2\u003e \u003cp\u003e This retrospective work received institutional approval, and the informed consent requirement was waived. In detail, ethical approval was obtained from the Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong First Medical University, China. In addition, patient informed consent was waived due to the retrospective nature of this study, which was under the permission of the Ethics Committee of Shandong Provincial Hospital Affiliated to Shandong First Medical University. All methods were carried out in accordance with relevant guidelines and regulations. Clinical trial number: not applicable.\u003c/p\u003e \u003cp\u003eFor this investigation, we recruited AAA patients via the medical/radiological reports of three separate hospitals between January 2014 and July 2022. The following patients were included in our analysis: (1) AAA patients who underwent two contrast-enhanced abdominal CT examinations with a time interval\u0026thinsp;\u0026gt;\u0026thinsp;6 months; and (2) no endovascular or open aortic repair before or during the aforementioned imaging interval. The following patients were excluded from analysis: (1) patients suffering from other aortic diseases that impacted image analysis, such as, aortic dissection, aortic occlusion, and so on; (2) missing clinical information; and (3) poor image quality.\u003c/p\u003e \u003cp\u003ePatient clinical information was retrieved from the hospital medical records, including patient age, sex, smoking history, hypertension, diabetes, hyperlipidemia, anti-hypertensive treatment, and lipid-lowering treatment.\u003c/p\u003e \u003cp\u003eIn all, 166 patients (147 men; average age, 69.2\u0026thinsp;\u0026plusmn;\u0026thinsp;10.2 years; 66 with AAA progression and 100 without AAA progression) were recruited from analysis. Among them, 92 patients (35 with AAA progression and 57 without AAA progression) from one hospital (Shandong Provincial Hospital) served as the training cohort. Additionally, 74 patients (31 with AAA progression and 43 without AAA progression) from the two remaining hospitals (Qilu Hospital of Shandong University; Shandong Provincial Qianfoshan Hospital) formed the external test cohort.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 CT imaging protocol\u003c/h2\u003e \u003cp\u003eAll contrast-enhanced abdominal CT images were acquired from the multidetector CT scanners (Somatom Force, Siemens Healthcare; Somatom Definition Flash, Siemens Healthcare; Ingenuity CT, Philips). Enhanced CT scanning encompassed the base of the neck to the aortoiliac bifurcation, with the patient positioned supine. Next, using a power injector, 90\u0026ndash;120 mL of contrast media (Omnipaque-350; GE Healthcare) was administered at 4-4.5 mL/s, before flushing with 40 mL saline at the same speed. Subsequently, bolus tracking was employed to acquire data at an attenuation cut-off of 100 Hounsfield units at the celiac trunk level for 6 s. Scanning was proceeded as follows: tube voltage: 120 kVp, pitch: 0.8-1.0, tube current: 300\u0026ndash;400 mA (via automatic tube current modulation), matrix: 512 \u0026times; 512, recreated slice thickness: 1 mm, recreated slice interval: 1 mm, rotation duration: 500\u0026ndash;600.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 CT imaging analysis and clinical model progression\u003c/h2\u003e \u003cp\u003eThe baseline CT imaging features were retrieved, including, maximum AAA diameter and total volume; with or without ILT; as well as maximum ILT diameter and volume. In addition, we also obtained the total AAA volume at the follow-up CT. ILT presence was described as ILT thickness, along with a\u0026thinsp;\u0026gt;\u0026thinsp;5 mm aortic wall on enhanced CT.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e Maximum AAA and ILT diameters were assessed based on the maximum cross-sections or maximum ILT cross-sections perpendicular to the long axis of the aorta using a multiplanar reconstruction method. The superior and inferior AAA borders were defined as the loss of aortic wall parallelism to the aortic dilatation end. The AAA progression rate was computed using the following formula: (AAA volume at follow-up - AAA volume at baseline) / follow-up years. AAA progression was defined as an annual AAA progression rate\u0026thinsp;\u0026ge;\u0026thinsp;10 mL/year.\u003csup\u003e8\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eAll imaging features were assessed and quantified by two cardiovascular radiologists (Z.S. and H.G., with 7 and 10 years of experience in vascular imaging, respectively). The evaluators were blinded to the subjects\u0026rsquo; clinical data using semiautomatic segmentation via the ITK-SNAP software (version 3.8.0, open source, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.itksnap.org\u003c/span\u003e\u003cspan address=\"http://www.itksnap.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe performed inter-patient comparisons of the clinical and CT features among participants with and without AAA progression using univariable analysis (quantitative data was assessed via independent-sample t tests, whereas, qualitative data via chi-square tests). Variables reaching significance in univariable analysis were entered into in multivariable logistic regression analysis to estimate AAA progression. Lastly, the significant independent AAA indicators from multivariable analysis were employed for clinical model construction.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Lesion segmentation and radiomics feature retrieval\u003c/h2\u003e \u003cp\u003eA flowchart depicting our radiomics analysis is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. A post-processing platform (Lianying Intelligent Medical Technology Co., Ltd., Beijing, China) was employed for lesion segmentation and radiomics feature acquisition. Prior to radiomics feature extraction, images were batch adjusted, namely, gray-scale discretization and image resampled, to manage differences in image quality and noise, and reduce radiomics features variability. Using manual segmentation, we next selected regions of interests (ROIs) in the lesion cross-sectional area. Contours were drawn within 5 mm of the aortic wall avoiding other organ tissues. Lastly, we extracted several radiomics variables, such as, the first-order statistics, shape, and textural features (gray-level size zone, gray-level co-occurrence, gray-level dependence, gray-level run-length, and neighborhood gray-tone differencematrix features).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo assess the inter- and intra-observer reproducibility of the aforementioned feature acquisitions, we employed the inter- and intra-class correlation coefficients (ICCs). 30 CT images (17 AAA progression and 13 AAA non-progression) were arbitrarily chosen for ROI segmentation by two independent radiologists (S.Z. and M.H.L.). Moreover, M.H.L. repeated the segmentation 3 weeks after the first acquisition to evaluate reproducibility of the extraction. An ICC\u0026thinsp;\u0026gt;\u0026thinsp;0.75 represented satisfactory agreement of feature acquisition. The remaining image segmentations were subsequently conducted by M.H.L, with proper blinding to patient clinical information.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Feature selection and radiomics signature and nomogram model progression\u003c/h2\u003e \u003cp\u003eWe first selected radiomics features with ICCs\u0026thinsp;\u0026gt;\u0026thinsp;0.75. Then, the remaining features were tested via the select_k_best method to identify potential significant features. Finally, the least absolute shrinkage and selection operator (LASSO) regression model was employed for the selection of the most significant features. Features identified in LASSO were utilized in radiomics signature model construction, and an radiomics score (Rad-score) was computed for individuals using a linear combination of identified features weighted by their corresponding LASSO coefficients. A radiomics-based nomogram was then established by incorporating the most relevant variables from the clinical model and Rad-score.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Statistical Analysis\u003c/h2\u003e \u003cp\u003eReceiver operating characteristic curve (ROC) analysis was employed for the measurement of the areas under the curve (AUCs) for predicting AAA progression in the training sets and external test sets of the clinical model, radiomics signature, and nomogram model. AUC comparisons were conducted via the Delong test. The predictive power of the nomogram model was assessed via the decision curve analysis (DCA). nomogram model calibration was performed using calibration curves. Two-sided \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was deemed as significant. SPSS (version 24.0, IBM) and R (version 3.3.3, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org\u003c/span\u003e\u003cspan address=\"https://www.r-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were employed for all data analyses.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.1 clinical model\u003c/h2\u003e \u003cp\u003eIn all, 166 individuals (147 men; average age, 69.2\u0026thinsp;\u0026plusmn;\u0026thinsp;10.2 years) were included in our analysis. The median follow-up duration was 0.94 years (ranging between 0.5\u0026ndash;7.05 years) with a progression rate of 16.9\u0026thinsp;\u0026plusmn;\u0026thinsp;28.8 mL/y. Among them, 92 patients (35 with AAA progression and 57 without AAA progression) served as the training set, and 74 patients (31 with AAA progression and 43 without AAA progression) as the external test set.\u003c/p\u003e \u003cp\u003eClinical and CT features of patients from both groups in the training set and external test set are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In the training set, AAA progression patients, relative to AAA non-progression patients, revealed enhanced diabetes incidences (42.9% vs. 12.3%, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001), larger AAA maximal diameter (44.2\u0026thinsp;\u0026plusmn;\u0026thinsp;5.3 mm vs. 39.2\u0026thinsp;\u0026plusmn;\u0026thinsp;6.4 mm, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and larger total aneurysm volume (79.8\u0026thinsp;\u0026plusmn;\u0026thinsp;66.9 mL vs. 51.4\u0026thinsp;\u0026plusmn;\u0026thinsp;44.1 mL, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03). The rest of the analyzed variables showed no obvious differences between the two subject populations (all \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05). Using multivariate analysis, we revealed that diabetes [odds ratio (OR): 3.77; 95% confidence interval (CI): 1.25\u0026ndash;11.42; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02] and AAA maximal diameter (OR: 1.11; 95% CI: 1.02\u0026ndash;1.20; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) were the independent predictors of AAA progression. Hence, these variables were employed for clinical model construction.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eClinical characteristics and CT imaging characteristics of patients with and without AAA progress\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eTraining Set (n\u0026thinsp;=\u0026thinsp;92)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eExternal test Set (n\u0026thinsp;=\u0026thinsp;74)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNon-progression Group (n\u0026thinsp;=\u0026thinsp;57)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eProgression Group (n\u0026thinsp;=\u0026thinsp;35)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNon-progression Group (n\u0026thinsp;=\u0026thinsp;43)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eProgression Group (n\u0026thinsp;=\u0026thinsp;31)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eP\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (y)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e69.3\u0026thinsp;\u0026plusmn;\u0026thinsp;9.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e68.1\u0026thinsp;\u0026plusmn;\u0026thinsp;8.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e68.5\u0026thinsp;\u0026plusmn;\u0026thinsp;12.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e71.6\u0026thinsp;\u0026plusmn;\u0026thinsp;9.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex, male\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e49 (86.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31 (88.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38 (88.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e29 (93.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41 (71.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26 (74.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.81\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e31 (72.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e21 (67.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7 (12.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (42.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12 (27.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9 (29.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHyperlipidemia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e24 (42.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (34.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16 (37.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8 (25.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking history\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28 (49.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19 (54.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30 (69.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e19 (61.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnti-hypertensive treatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e44 (77.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28 (80.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33 (76.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e23 (74.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLipid-lowering treatment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26 (45.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (37.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e19 (44.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e10 (32.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaximal diameter (mm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39.2\u0026thinsp;\u0026plusmn;\u0026thinsp;6.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e44.2\u0026thinsp;\u0026plusmn;\u0026thinsp;5.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e39.7\u0026thinsp;\u0026plusmn;\u0026thinsp;6.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e44.3\u0026thinsp;\u0026plusmn;\u0026thinsp;6.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal aneurysm volume (mL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e51.4\u0026thinsp;\u0026plusmn;\u0026thinsp;44.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e79.8\u0026thinsp;\u0026plusmn;\u0026thinsp;66.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e65.2\u0026thinsp;\u0026plusmn;\u0026thinsp;58.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e101.2\u0026thinsp;\u0026plusmn;\u0026thinsp;77.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresence of ILT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42 (73.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31 (88.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e31 (72.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22 (71.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eILT maximal diameter (mm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8.5\u0026thinsp;\u0026plusmn;\u0026thinsp;7.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.2\u0026thinsp;\u0026plusmn;\u0026thinsp;7.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11.3\u0026thinsp;\u0026plusmn;\u0026thinsp;9.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8.7\u0026thinsp;\u0026plusmn;\u0026thinsp;8.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eILT volume (mL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.5\u0026thinsp;\u0026plusmn;\u0026thinsp;28.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33.5\u0026thinsp;\u0026plusmn;\u0026thinsp;44.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e26.2\u0026thinsp;\u0026plusmn;\u0026thinsp;53.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22.4\u0026thinsp;\u0026plusmn;\u0026thinsp;26.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eContinuous variables are described as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, and categorical variables are presented as numbers (%). AAA, abdominal aortic aneurysm; ILT, Intraluminal thrombus.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Development of radiomics signature and nomogram model\u003c/h2\u003e \u003cp\u003eOverall, 2264 radiomics features were obtained from the CT images. Among them, 1412 were selected following elimination of features with ICC\u0026thinsp;\u0026lt;\u0026thinsp;0.75. Out of the 1412, 313 revealed obvious differences between the two patient populations (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) as evidenced by the select_best_k method. Among them, 9 radiomics features were selected as the most relevant, using LASSO. Radiomics signature was generated based on these 9 radiomics features (termed A-I) and their matched LASSO coefficients as follows: 0.40\u0026thinsp;+\u0026thinsp;0.074 \u0026times; A\u0026thinsp;+\u0026thinsp;0.069 \u0026times; B\u0026thinsp;+\u0026thinsp;0.046 \u0026times; C\u0026thinsp;+\u0026thinsp;0.044 \u0026times; D\u0026thinsp;+\u0026thinsp;E \u0026times; 0.042\u0026thinsp;+\u0026thinsp;F \u0026times; 0.038 - G \u0026times; 0.033 - H \u0026times; 0.038 - I \u0026times; 0.061 (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Out of the 9 radiomics features, three exhibited the highest coefficients, and they were shape_maximum2ddiameterslice, firstorder_wavelet-hll-kurtosis, and firstorder_wavelet-hlh-skewness. In the external test set, the shape_maximum 2d diameterslice (64.9\u0026thinsp;\u0026plusmn;\u0026thinsp;18.1 vs. 55.2\u0026thinsp;\u0026plusmn;\u0026thinsp;9.1; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.004), firstorder_wavelet-hll-kurtosis (30.2\u0026thinsp;\u0026plusmn;\u0026thinsp;13.5 vs. 20.5\u0026thinsp;\u0026plusmn;\u0026thinsp;14.7; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.005), and firstorder_wavelet-hlh-skewness (0.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3 vs. 0.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3; \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.009) were obviously enhanced among AAA progression patients, compared to the AAA non-progression patients (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRadiomics features selected for inclusion in radiomics signature.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFeature Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFeature\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLASSO Coefficient\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShape\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003emaximum2ddiameterslice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.074\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFirst order\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ewavelet-hll-kurtosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.069\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGLSZM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elargearealowgraylevelemphasis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.046\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFirst order\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNormalize mean\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.044\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGLSZM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003elog-sigma-2-mm-3d-smallareahighgraylevelemphasis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFirst order\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ewavelet-llh-totalenergy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eG\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGLSZM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBoxsigmaimage zonepercentage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.033\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eShape\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003esphericity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.038\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFirst order\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ewavelet-hlh-skewness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.061\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eGLSZM\u0026thinsp;=\u0026thinsp;gray level size zone matrix, LASSO\u0026thinsp;=\u0026thinsp;least absolute shrinkage and selection operator.\u003c/p\u003e \u003cp\u003eMoreover, the radiomics signature was considerably high among AAA progression patients, compared to AAA non-progression patients within the training set (2.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7 vs. 1.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and external test set (1.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9 vs. 1.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eThe AAA progression-predicting nomogram model was generated by incorporating diabetes, AAA maximal diameter, and radiomics signature (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Based on the calibration curves, this nomogram model exhibited satisfactory calibration in the training set and external test set (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Performance evaluation\u003c/h2\u003e \u003cp\u003eThe predictive abilities of clinical model, radiomics signature, and nomogram model in diagnosing AAA progression in the training set and external test set are presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Inter-AUC differences among models are also provided in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Lastly, the ROC curves of all models in the training set and external test set are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDiagnostic performance of clinical model, radiomics signature, and nomogram\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMeasure\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eClinical Model\u003c/p\u003e \u003cp\u003e[1]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eRadiomics Signature\u003c/p\u003e \u003cp\u003e[2]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eNomogram\u003c/p\u003e \u003cp\u003e[3]\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1 vs 2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1 vs 3\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2 vs 3\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e72.8 (67/92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e77.2 (71/92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e83.7 (77/92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.9 (29/35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e80.0 (28/35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.1 (27/35)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e66.7 (38/57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e75.4 (43/57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e87.7 (50/57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExternal test set\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAccuracy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e68.9 (51/74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e71.6 (53/74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e78.4 (58/74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e45.2 (14/31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e83.9 (26/31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e74.2 (23/31)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e86.1 (37/43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62.8 (27/43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e81.4 (35/43)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eNote\u0026mdash;Accuracy, sensitivity, and specificity reported as percentages with numerators and denominators in parentheses. AUC, Area under the curve.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the training set, the AUCs were 0.79, 0.86, and 0.89 for the clinical model, radiomics signature, and nomogram model, respectively. The corresponding accuracy, sensitivity, and specificity for clinical model were 72.8% (67 of 92 patients), 82.9% (29 of 35 patients), and 66.7% (38 of 57 patients), respectively; for radiomics signature, 77.2% (71 of 92 patients), 80.0% (28 of 35 patients), and 75.4% (43 of 57 patients), respectively; and for nomogram model, 83.7% (77 of 92 patients), 77.1% (27 of 35 patients), and 87.7% (50 of 57 patients), respectively.\u003c/p\u003e \u003cp\u003eThe nomogram model AUC was considerably higher, compared to the clinical model AUC (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02). However, no obvious differences were observed in the AUCs between radiomics signature and those of clinical model and nomogram model (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.26 and \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.20, respectively).\u003c/p\u003e \u003cp\u003eIn the external test set, the AUCs were 0.69, 0.77, and 0.83 for the clinical model, radiomics signature, and nomogram model, respectively. The accuracy, sensitivity, and specificity for clinical model were 68.9% (51 of 74 patients), 45.2% (14 of 31 patients), and 86.1% (37 of 43 patients), respectively; for radiomics signature were 71.6% (53 of 74 patients), 83.9% (26 of 31 patients), and 62.8% (27 of 43 patients), respectively; and for nomogram model were 78.4% (58 of 74 patients), 74.2% (23 of 31 patients), and 81.4% (35 of 43 patients), respectively. The nomogram model AUC was considerably higher, compared to the clinical model AUC (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02). However, no obvious differences were observed in the AUCs between radiomics signature and those of clinical model and nomogram model (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.36 and \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.12, respectively). The DCAs of all models are provided in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBased on our analysis, relative to clinical model, nomogram model exhibited superior overall benefit in delineating between AAA progression and AAA non-progression patients over a wide range of cut-off possibilities at which an AAA progression diagnosis is likely to occur. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e illustrates the CT images of representative patients with and without AAA progression, and their corresponding nomogram results.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eAAA is a widespread disease with high mortality\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Hence, the accurate prediction of AAA progression is vital to the clinical management and intervention of this disease\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Herein, we established and verified a CT-guided nomogram of AAA progression estimation by incorporating clinical model and radiomics signature data. The nomogram model demonstrated satisfactory diagnostic performance in the training set (AUC, 0.83; sensitivity, 74.2%; specificity, 81.4%; and accuracy, 78.4%), and was superior to clinical model (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) in the external test set. These findings suggested that the clinical-radiomics nomogram is an effective tool for predicting AAA progression in the clinics.\u003c/p\u003e \u003cp\u003eThe AAA diameter is a widely investigated image-based predictive factor that is closely associated with progression and rupture, and is the main indication for treatment\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Several earlier studies revealed that a larger initial diameter results in larger expansion rates and is intricately linked to AAA progression\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. This is consistent with the conclusions of our study, whereby we demonstrated that the AAA maximal diameters (OR: 1.11; 95% CI: 1.02\u0026ndash;1.20; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) were independent indicators of AAA progression. Moreover, we identified that diabetes (OR: 3.774; 95% CI: 1.25\u0026ndash;11.42; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.02) was another independent indicator of AAA progression. A possible reason for this is that hyperglycemia damages the inner blood vessel wall by increasing local inflammatory factor and free radical accumulations, which, in turn, inflicts damage to the vascular endothelial structure and overall integrity of the blood vessel wall, thereby promoting AAA progression\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Several reports demonstrated a strong relation between ILT and AAA progression, and revealed that the ILT size and volume are strongly linked to AAA occurrence\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Interestingly, herein, we did not observe an obvious difference in the ILT diameter and volume between the progressive and non-progressive cohorts. The reason for this difference may be our relatively smaller sample size.\u003c/p\u003e \u003cp\u003eArtificial intelligence analysis is an emerging noninvasive tool for the extraction of quantitative features from images, thus compensating for the limitations of visual image evaluation\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. At present, there is growing interest in research involving the detection of disease, cardiovascular events prediction analysis, and assessment of prognosis in cardiovascular radiology\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Lee et al.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e examined 94 AAA patients using machine learning to predict AAA progression. Moreover, Wang et al.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e established models for estimating prognosis following endovascular AAA repair based on morphology, deep learning, and radiomics signature. Herein, we generated a clinical-radiomics nomogram model to predict AAA progression, and our model displayed satisfactory diagnostic performance (AUC: 0.83, in external test set).\u003c/p\u003e \u003cp\u003eHerein, the radiomics feature with the highest LASSO coefficient was Shape_maximum2ddiameterslice, which reflected lesion morphology. This was similar to the clinical model in our study, whereby we concluded that the AAA progression patients possessed larger AAA maximal diameters than non-AAA progression patients. Thus, diameter was a major parameter predicting AAA progression. However, the machine automatically measured Shape_maximum2ddiameterslice on horizontal images, whereas, clinical model measurements were made in the cross-sections perpendicular to the long aortic axis using a multiplanar reconstruction method. Moreover, the radiomics features firstorder_wavelet-hll-kurtosis and firstorder wavelet hlh skewness were significantly enhanced among AAA progression patients, suggesting greater inhomogeneity of image pixels\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. This difference may be attributed to the greater blood flow complexity or heterogeneity of surrounding tissue among AAA progression patients. In addition, several studies revealed that the computational fluid dynamics parameters and perivascular inflammation may also predict AAA progression\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHerein, we selected the CT angiography (CTA) as our assessment modality. Relative to ultrasound, CTA employs three-dimensional reconstruction, which greatly reduces subjectivity influence \u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. In addition, compared to the magnetic resonance imaging examination, CTA scanning provides faster results, and avoids MRI contraindications, such as, claustrophobia, cardiac pacemaker, and so on\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Given that the contrast agent may conceal the textural data of parts of images during CTA examination, we acquired\u0026thinsp;\u0026gt;\u0026thinsp;5 mm area around the AAA. Several investigations demonstrated that 5 mm of adipose tissue surrounding the AAA reflects the degree of aneurysm-based inflammation, which eventually influences AAA progression\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Multiple studies employed alterations in the vessel diameter to define AAA progression\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Although vessel diameter is easy to measure, the volume is more sensitive to alterations than diameter\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Therefore, similar to other studies, we employed alterations in the vessel volume to define AAA progression.\u003c/p\u003e \u003cp\u003eThis investigation has certain limitations. First, being a retrospective study, there may be certain unintentional selection bias. Second, despite being a multicenter study, our sample size was relatively small. Thus, we recommend future investigations with a larger patient population and longer follow-up duration. Third, in an attempt to minimize radiation from plain scan, most patients only underwent CTA examination. However, given that the contrast agent may conceal the textural information of images, a plain scan would have provided a certain research value, and would be the content of our next investigation. Forth, CT features were not examined at follow-ups, and there was no assessment of alterations between pre- and post-CT images. We consider that the baseline CT features are more important in predicting AAA progression, and our established model constructed with the baseline CT feature demonstrated satisfactory diagnostic efficacy.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eIn conclusion, herein, we demonstrated that the nomogram model, which incorporated data from the clinical model and radiomics signature, displayed superior diagnostic performance than the clinical model. Based on our conclusions, the CT-based clinical-radiomics nomogram may be a robust tool for the prediction of AAA progression, and may accurately guide clinical management in the near future.\u003c/p\u003e "},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAAA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAbdominal aortic aneurysm\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArea under the curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eConfidence interval\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCTA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCT angiography\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDCA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDecision curve analysis\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eICC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eInter and intra-class correlation coefficients\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eILT\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eIntraluminal thrombus\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLASSO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLeast absolute shrinkage and selection operator\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eOdds ratio\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRegions of interest.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThe present study was supported by the National Natural Science Foundation of China (grant no. 81871354, 81571672), and Academic promotion programme of Shandong First Medical University (grant no. 2019QL023).\u003c/p\u003e\n\u003cp\u003eDeclaration of Competing Interest\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003cp\u003eContributions\u003c/p\u003e\n\u003cp\u003eSZ and XMW were involved in conceptualization, writing-original Draft, writing-review and editing; RT was involved in methodology, data Curation, and formal analysis; MBW and BK were involved in formal analysis and funding acquisition; XXY and GHZ involved in investigation, data curation and project administration; MBW involved in methodology and software and validation; XMW involved in resources, visualization, supervision and funding acquisition. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eThe data used and analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003eConflict of Interest Statement\u003c/p\u003e\n\u003cp\u003eThe authors have no conflicts of interest to declare.\u003c/p\u003e\n\u003cp\u003eConsent to Publish declaration:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eNone.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSchanzer A, Oderich GS. Management of Abdominal Aortic Aneurysms. N Engl J Med. 2021;385:1690\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUS Preventive Services Task Force, Owens DK, Davidson KW, et al. Screening for Abdominal Aortic Aneurysm: US Preventive Services Task Force Recommendation Statement. JAMA. 2019;322:2211\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuirguis-Blake JM, Beil TL, Senger CA, Coppola EL. Primary Care Screening for Abdominal Aortic Aneurysm: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2019;322:2219\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollaborators RESCAN, Bown MJ, Sweeting MJ, Brown LC, Powell JT, Thompson SG. Surveillance intervals for small abdominal aortic aneurysms: a meta-analysis. JAMA. 2013;309:806\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKosmas I, Paraskevas A, Mansilha. Implications of abdominal aortic aneurysm rupture at a lower diameter than the recommended threshold for AAA repair. Int Angiol. 2023;42(4):279\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChaikof EL, Dalman RL, Eskandari MK, et al. The Society for Vascular Surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J Vasc Surg. 2018;67:2\u0026ndash;e772.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu C, Leach JR, Wang Y, Gasper W, Saloner D, Hope MD. Intraluminal Thrombus Predicts Rapid Growth of Abdominal Aortic Aneurysms. Radiology. 2020;294:707\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeyrignac O, Bal L, Zadro C, et al. Combining Volumetric and Wall Shear Stress Analysis from CT to Assess Risk of Abdominal Aortic Aneurysm Progression. Radiology. 2020;295:722\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFan YN, Ke X, Yi ZL, et al. Plasma D-dimer as a predictor of intraluminal thrombus burden and progression of abdominal aortic aneurysm. Life Sci. 2020;240:117069.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRauschecker AM, Rudie JD, Xie L, et al. Artificial Intelligence System Approaching Neuroradiologist-level Differential Diagnosis Accuracy at Brain MRI. Radiology. 2020;295:626\u0026ndash;37.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKosmas I, Paraskevas; Armando M. Implications of abdominal aortic aneurysm rupture at a lower diameter than the recommended threshold for AAA repair. Int Angiol. 2023;42(4):279\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong Z, Tang Z, Liu H, Guo D, Cai J, Zhou Z. A clinical-radiomics nomogram may provide a personalized 90-day functional outcome assessment for spontaneous intracerebral hemorrhage. Eur Radiol. 2021;31:4949\u0026ndash;59.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y, Zhou M, Ding Y, et al. Development and Comparison of Multimodal Models for Preoperative Prediction of Outcomes After Endovascular Aneurysm Repair. Front Cardiovasc Med. 2022;9:870132.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing N, Hao Y, Wang Z, et al. CT texture analysis predicts abdominal aortic aneurysm post-endovascular aortic aneurysm repair progression. Sci Rep. 2020;10:12268.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHirata K, Nakaura T, Nakagawa M, et al. Machine Learning to Predict the Rapid Growth of Small Abdominal Aortic Aneurysm. J Comput Assist Tomogr. 2020;44:37\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee R, Jarchi D, Perera R, et al. Applied Machine Learning for the Prediction of Growth of Abdominal Aortic Aneurysm in Humans. EJVES Short Rep. 2018;39:24\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGandhi R, Bell M, Bailey M, Tsoumpas C. Prospect of positron emission tomography for abdominal aortic aneurysm risk stratification. J Nucl Cardiol. 2021;28:2272\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNana P, Spanos K, Dakis K, Brodis A, Kouvelos G. Imaging Predictive Factors of Abdominal Aortic Aneurysm Growth. \u003cem\u003eJ Clin Med\u003c/em\u003e. 2021; 10:1917.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePetroula Nana K, Spanos K, Dakis, et al. Imaging Predictive Factors of Abdominal Aortic Aneurysm Growth. J Clin Med. 2021;10(9):1917\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBel\u0026eacute;n Picatoste I, Cerro-Pardo LM, Blanco-Colio, et al. Protection of diabetes in aortic abdominal aneurysm: Are antidiabetics the real effectors? Front Cardiovasc Med. 2023;10(0):1112430\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYingNan F, Ke X, Yi ZL, et al. Plasma D-dimer as a predictor of intraluminal thrombus burden and progression of abdominal aortic aneurysm. Life Sci. 2020;240(0):117069\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiangliang Zhang, Byron A, Zambrano J, Choi, et al. Intraluminal thrombus effect on the progression of abdominal aortic aneurysms by using a multistate continuous-time Markov chain model. J Int Med Res. 2020;48(11):300060520968449.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo Y, Chen X, Lin X, et al. Non-contrast CT-based radiomic signature for screening thoracic aortic dissections: a multicenter study. Eur Radiol. 2021;31:7067\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu W, Wu X, Dong D, et al. Novel radiomics features from CCTA images for the functional evaluation of significant ischaemic lesions based on the coronary fractional flow reserve score. Int J Cardiovasc Imaging. 2020;36:2039\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeng C, Yang W, Lan H, Ren X, Li M. Development and Application of a Vehicle-Mounted Soil Texture Detector. Sens (Basel). 2020;20:7175.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDias-Neto M, Meekel JP, van Schaik TG, et al. High Density of Periaortic Adipose Tissue in Abdominal Aortic Aneurysm. Eur J Vasc Endovasc Surg. 2018;56:663\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTimothy P. Szczykutowicz. Computed Tomography Angiography: Principles and Advances. Radiol Clin North Am. 2024;62(3):371\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChiara Stassi C, Mondello G, Baldino, et al. State of the Art on the Role of Postmortem Computed Tomography Angiography and Magnetic Resonance Imaging in the Diagnosis of Cardiac Causes of Death: A Narrative Review. Tomography. 2022;8(2):961\u0026ndash;73.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLe S, Wu J, Liu H, et al. Single-cell RNA sequencing identifies interferon-inducible monocytes/macrophages as a cellular target for mitigating the progression of abdominal aortic aneurysm and rupture risk. Cardiovasc Res. 2024;120(11):1351\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDrew J, Braet TJ, Baker L, Delbono, et al. Three-dimensional characterization of sex differences in abdominal aortic aneurysm progression via vascular deformation mapping. Sci Rep. 2024;14(1):24215.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBouwens E, Vanmaele A, Sanne E, Hoeks, et al. Circulating biomarkers of cardiovascular disease are related to aneurysm volume in abdominal aortic aneurysm. Vasc Med. 2023;28(5):433\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Abdominal aortic aneurysm, Radiomics, Aneurysm progression, computer tomography angiography, nomogram ","lastPublishedDoi":"10.21203/rs.3.rs-6439190/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6439190/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e The rapid progression associated with abdominal aortic aneurysms (AAA) is the major contributor to morbidity and mortality among the elderly. More comprehensive radiomics-based prediction of AAA progression is both valuable and warranted.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e In this multicenter retrospective investigation, 166 AAA patients who received two contrast-enhanced abdominal CT examinations from January 2014 to July 2022 were divided into training (n = 92) and external test cohort (n = 74). A clinical model for predicting AAA progression was built using clinical and CT characteristics that were significant independent predictors. Radiomics features were extracted from CT images, and a radiomics signature was constructed. The nomogram model was constructed by combining clinical model and radiomics signature. The diagnostic performance were evaluated and validated on the training and test sets, and then compared among the three models.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Over a median of 0.94 years (range, 0.5-7.05 years), 66 patients (training set: n = 35; external test set: n= 31) experienced AAA progression. The clinical model was built by diabetes and AAA maximal diameter. Seven features were employed to build the radiomics signature. In the external test set, the area under the curve was higher for the nomogram (0.83) than for the clinical model (0.69, \u003cem\u003ep\u003c/em\u003e =0.02), and the nomogram model showed a sensitivity, specificity, and accuracy of 74.2%, 81.4%, and 78.4%.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e The nomogram model combining the clinical factors and radiomics signature showed better diagnostic performance than the clinical model, and may assist clinical decision-making process.\u003c/p\u003e","manuscriptTitle":"Development and validation of CT-based clinical-radiomics nomogram for predicting abdominal aortic aneurysms progression","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-20 09:39:53","doi":"10.21203/rs.3.rs-6439190/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-05T03:09:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"154174855620888532240496422035755011735","date":"2026-04-26T11:31:09+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-05-18T14:05:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"38158843776876438330924749357351626532","date":"2025-05-18T10:21:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-15T03:25:48+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-04-22T09:00:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-19T11:28:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-19T11:27:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Informatics and Decision Making","date":"2025-04-13T12:28:06+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"80d46780-017b-4239-8404-75856d6c3f3a","owner":[],"postedDate":"May 20th, 2025","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-05T03:09:44+00:00","index":76,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-05-20T09:39:53+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-20 09:39:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6439190","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6439190","identity":"rs-6439190","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}