Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model

Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model | 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 Article Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model Zhanwei Fu, Shuchun Li, Lu Zang, Feng Dong, Zhenghao Cai, Junjun Ma This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2681419/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 02 Nov, 2023 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Background Multiple linear stapler firings is a risk factor for anastomotic leakage (AL) in laparoscopic low anterior resection (LAR) using double stapling technique (DST) anastomosis. In this study, our objective was to establish the risk factors for ≥3 linear stapler firings, and to create and validate a predictive model for ≥3 linear stapler firings in laparoscopic LAR using DST anastomosis. Methods We retrospectively enrolled 328 mid–low rectal cancer patients undergoing laparoscopic LAR using DST anastomosis. With a split ratio of 4:1, patients were randomly divided into 2 sets: the training set (n = 260) and the testing set (n = 68). A clinical predictive model of ≥3 linear stapler firings was constructed by binary logistic regression. Based on three-dimensional convolutional networks, we built an image model using only magnetic resonance (MR) images segmented by Mask region-based convolutional neural network, and an integrated model based on both MR images and clinical variables. Area under the curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV), and Youden index were calculated for each model. And the three models were externally validated by an independent cohort of 128 patients. Results There were 17.7% (58/328) patients received ≥3 linear stapler firings. Tumor size ≥5 cm (odds ratio (OR)=2.54, 95% confidence interval (CI)=1.15–5.60, p=0.021) and preoperative carcinoma embryonic antigen (CEA) level >5 ng/mL [OR=2.20, 95% CI=1.20–4.04, p=0.011] were independent risk factors associated with ≥3 linear stapler firings. The integrated model (AUC=0.88, accuracy=94.1%) performed better on predicting ≥3 linear stapler firings than the clinical model (AUC =0.72, accuracy=86.7%) and the image model (AUC=0.81, accuracy=91.2%). Similarly, in the validation set, the integrated model (AUC=0.84, accuracy=93.8%) performed better than the clinical model (AUC =0.65, accuracy=65.6%) and the image model (AUC=0.75, accuracy=92.1%). Conclusion Our deep-learning model based on pelvic MR can help predict the high-risk population with ≥3 linear stapler firings in laparoscopic LAR using DST anastomosis. This model might assist in determining preoperatively the anastomotic technique for mid–low rectal cancer patients. Biological sciences/Cancer Health sciences/Gastroenterology Health sciences/Medical research Low anterior resection (LAR) Double stapling technique (DST) Rectal cancer Magnetic resonance imaging (MRI) Deep-learning Predictive model Figures Figure 1 Figure 2 Figure 3 Introduction Colorectal cancer (CRC) is the third most prevalent cancer worldwide with high rate of cancer mortality, while approximately one third of all CRCs occur in rectum [1] . Total mesorectal excision (TME) is the primary treatment for localized rectal cancer patients, and has high rate of cancer control [2, 3] . It is a widely applied surgical technique to perform laparoscopic low anterior resection (LAR) using double-stapling technique (DST) anastomosis [4, 5] . In mid–low rectal cancer patients, anastomotic leakage (AL) is the most common postoperative complication after LAR [6] . In addition to increasing the incidence of reoperation and mortality, AL may even negatively affect long-term survival [7-9] . Despite advances in surgical techniques and anastomotic equipment over the past few decades [10-12] , there has been no significant decrease in the incidence of AL after LAR [13] . DST simplifies colorectal reconstruction in LAR, especially in laparoscopic surgery, but the incidence of postoperative AL was not reduced by the application of this technique [14, 15] . Several studies have proved that ≥3 linear stapler firings in laparoscopic LAR is an independent risk factor for AL [16-18] . Thus, the Chinese expert consensus statement proposes ≤2 linear stapler firings in LAR surgery [19] . To avoid multiple linear stapler firings, a predictive model should be created to predict ≥3 linear stapler firings in anastomosis, and alternative techniques could be considered. Several studies have shown that pelvic anatomical features (such as the anteroposterior diameter and the transverse diameter of the pelvic outlet, the anteroposterior diameter of the pelvic inlet) are related to surgical difficulty and the number of the linear stapler firings [20-22] . Those studies mainly measured pelvic parameters, but did not include the influence of rectal and mesenteric conditions. How to effectively integrate above parameters to predict ≥3 linear stapler firing needs further research. With the advancement of imaging technology, pelvic magnetic resonance imaging (MRI) is a preferred tool for local staging of mid-low rectal cancer before surgery [23, 24] . MRI can obtain relevant parameters of tumor, meso-rectum and pelvis comprehensively and accurately. Mask region-based convolutional neural network (Mask R-CNN) [25] and three-dimensional convolutional network (C3D) [26] image recognition are current techniques for advanced-recognition artificial intelligence (AI), and have been applied in various medical fields [27-29] . With advanced deep-learning technology, pelvic MRIs’ complex data can be identified, extracted, analyzed, and integrated efficiently, and we can create a predictive model to screen out high-risk patients with ≥3 linear stapler firings based on the database. In this study, we aimed to establish the risk factors for ≥3 linear stapler firings in laparoscopic LAR using DST anastomosis, and to create and validate a predictive model for ≥3 linear stapler firings with deep-learning technology based on MRI. Methods Patients A total of 328 mid–low rectal cancer patients who underwent laparoscopic LAR at Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine, China, between January 2016 and June 2021 were retrospectively analyzed as the deep-learning set. Clinical data were obtained from Ruijin hospital database and medical records. All methods were performed in accordance with the relevant guidelines and regulations, and the study was approved by the Medical Ethics Committee of Ruijin Hospital (No. 2019-82). The need for informed consent was waived by Ethics Committee of Ruijin Hospital. With a split ratio of 4:1, 260 patients were divided into the training set and 68 patients were divided into the testing set on the basis of an unbiased random sampling method. The external validation set comprised 128 patients from an independent clinical trial in our institution (Transanal versus laparoscopic total mesorectal excision for rectal cancer, ClinicalTrials.gov Identifier: NCT03359616). The inclusion criteria were: (1) histopathologically confirmed rectal cancer; (2) pelvic MRI examination within 14 days before surgery; (3) tumor distance from the anal verge ≤10 cm; and (4) laparoscopic LAR with DST anastomosis. The exclusion criteria were: (1) LAR without anastomosis (e.g., Hartmann’s operation); (2) anastomotic techniques rather than DST (e.g., manual rectal anastomosis through the anus); (3) number of linear stapler firings was not recorded on surgical reports; and (4) robotic surgery. Surgical procedure Laparoscopic LAR was performed by the gastrointestinal surgery team with experience in completing more than 100 rectal cancer operations every year. The laparoscopic LAR surgical procedure was carried out in strict accordance with the national guidelines for laparoscopic radical resection of CRC (2018 edition). During dissection of the distal rectum, the surgeon manually fired endoscopic linear staplers (Endo-GIA TM Ultra Universal Stapler Reload with Tri-stapler TM Technology; Covidien LLC), which was loaded with 60- or 45-mm staple cartridges which have three types of heights: 3.0, 3.5 and 4.0 mm. Data collection and model building We collected clinical data that potentially correlated with the number of linear staplers used in surgery, including baseline characteristics, including age, gender and body mass index (BMI); biochemical data, including hemoglobin, albumin, and carcinoma embryonic antigen (CEA); and tumor characteristics, including tumor distance from the anal verge, tumor stage, tumor size and circumferential resection margin (CRM). A predictive model of ≥3 linear stapler firings was constructed by binary logistic regression. The variables of the clinical model included: gender, BMI, serum CEA level (> or ≤5 ng/mL), tumor distance from the anal verge, tumor size and CRM. MRI and target area labeling During MRI, patients were in the supine position and scanned using a Philips INGENIA TM scanner with 3.0 T field strength. The pelvic phased-array surface coil covered from the aortic bifurcation to the anal verge. The scanning parameters were: layer thickness 5 mm; field of view 250´340´166 mm; echo time 80 ms; repetition time 3565 ms; and image matrix 312´357. With the Picture Archiving and Communication System, fat-suppressed fast spin-echo (FSE) T2-weighted sequences in the axial plane of the pelvis were used for image segmentation. Then, pelvic MRI specialists who have >10 years of experience built an image database by an online annotation tool called Labelme (labelme.csail.mit.edu/) [30] , and labeled three kinds of target area on each of the T2-weighted images (tumor body, mesorectum, and pelvis represented by green, yellow, and drab, respectively, Figure 1A-B). Then, all data were transformed into a COCO dataset for segmentation experiments [31] . Segmentation model based on Mask R-CNN As an effective two-stage detection and segmentation algorithm, Mask R-CNN was adopted to identify and segment the three kinds of target areas in the image (Figure 1C-D). In the first stage, the ResNet-101-FPN network served as the backbone to extract multiscale and discriminative feature maps. The Region Proposal Network scanned the feature map in a sliding-window and selected the rough detection rectangle that contained the object. After the regions of interest alignment process, the candidate regions entered the next stage. This consisted of three functional branches: classification, detection and segmentation, based on fully connected layers and convolutional layers. We trained the Mask R-CNN network on the training set for 200 epochs, and evaluated the performance of the testing set with standard COCO metrics. We evaluated the trained Mask R-CNN model, obtained average precision (AP) by calculating the precision–recall curve under different intersection-over-union thresholds, and then calculated the three types of targets. The respective AP values of the regions were weighted to obtain the class-wide mean average precision (mAP). mAP>50 indicated that the model performed well [32] . Deep-learning model based on C3D We used C3D networks to address 24 images of a case simultaneously and learn 3D spatial features. The C3D network consisted of eight 3D-convolution layers, a softmax layer, two fully connected layers and five pooling layers. It took the entire image of the case as input and output the probability of ≥3 linear stapler firings, and the sample with probability >50% (empirical value) was judged as positive. The C3D network was trained until convergence (~1000 epochs) and evaluated the performance of the deep-learning model. In the training set, two C3D-based models were trained, including an image model using only MR images and an integrated model based on both MR images and the six clinical variables in clinical model. In the testing set, the clinical model, image model and integrated model were examined, and receiver operating characteristic (ROC) curves were plotted. Area under the curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV), and Youden index were calculated. External validation We used clinical data and T2-weighted images of patients from the validation set to externally validate the predictive performance of the above three models. ROC curves were plotted, and sensitivity, specificity, accuracy, Youden index, PPV, and AUC were calculated. The flow chart of the design is shown in Figure 2. Statistical analysis Statistical analysis was performed using SPSS (version 25.0). Categorical variables were analyzed by Fisher’s exact test or Pearson’s chi-square. In our binary logistic regression models, only factors with a P value<0.10 in the univariate analysis were entered into the multivariate analysis. All tests were two-sided, and differences were considered statistically significant at p<0.05. Results Clinical characteristics of patients in deep-learning set The 328 patients had a median age of 63 (24–87) years, including 227 men and 101 women. The proportion of patients who received ≥3 linear stapler firings was 17.7% (58/328). Clinical characteristics of patients with ≥3 firings of the linear stapler and those with ≤2 firings in the deep-learning set were compared (Table 1). There were no significant differences in age, gender, BMI, diabetes mellitus, neoadjuvant chemoradiotherapy ratio, operation time, hemoglobin, albumin, tumor distance from the anal verge, T stage and N stage between the two groups. Patients with ≥3 firings showed significantly higher incidence of AL than those with ≤2 firings (p=0.021), and patients with ≥3 firings of the linear stapler showed significantly higher CEA level (p=0.007), larger tumor size (p=0.004) and higher rate of positive CRM (p=0.014). In univariate and multivariate analyses, tumor size ≥5 cm (odds ratio (OR)=2.54, 95% confidence interval (CI)=1.15–5.60, p=0.021) and serum CEA >5 ng/mL [OR=2.20, 95% CI=1.20–4.04, p=0.011] were independent risk factors associated with ≥3 linear stapler firings (Table 2). Predicting performance in testing set The AUCs of the clinical, imaging and integrated models were obtained as 0.72, 0.81 and 0.88, respectively (Figure 3A–C). The sensitivity, specificity, accuracy, PPV and Youden index of the clinical model were: 70.0%, 81.0%, 79.4%, 38.9%, and 0.51, respectively. The relevant indicators of the image model were: 50.0%, 98.3%, 91.2%, 83.3%, and 0.48, respectively. The relevant indicators of the integrated model were: 70.0%, 98.3%, 94.1%, 87.5%, and 0.68, respectively. External validation In the validation set, the proportion of patients who received ≥3 linear stapler firings was 12.5% (16/128). All related clinical characteristics between the deep-learning and validation sets were comparable (Table 3). The AUCs of the clinical, imaging and integrated models were obtained as 0.65, 0.75 and 0.84, respectively (Figure 3D–F). The sensitivity, specificity, accuracy, PPV and Youden index of the clinical model were: 62.5%, 66.1%, 65.6%, 21.0%, and 0.29, respectively. The relevant indicators of the image model were: 68.8%, 95.5%, 92.1%, 68.8%, and 0.64, respectively. The relevant indicators of the integrated model were: 68.8%, 97.3%, 93.8%, 78.5%, and 0.66, respectively. Discussion LAR using DST anastomosis is currently a widely applied surgical technique for mid–low rectal cancer, a series of high-quality randomized controlled trials has confirmed its feasibility and safety [5, 33] . The technique greatly reduces the difficulty of reconstruction of the digestive tract. However, some studies have reported that multiple linear stapler firings is closely related to AL [17, 34] , and AL is more likely to occur at the intersection of two staples [35] . In some cases, due to the limitation of the pelvic space or thickness of the rectum, the surgeons have to trigger more linear stapler firings during rectal dissection [36] . In the high-risk populations with ≥3 linear stapler firings, other anastomosis techniques rather than DST anastomosis could be considered, such as transanal anastomosis after transanal transection of the rectum [37] and manual purse-string suture after endoluminal transection of the rectum (e.g., Transanal total mesorectal excision) [38] . Although some studies have shown that the above techniques do not reduce the incidence of AL [39] , these techniques can minimize the anastomotic difficulty in patients with a narrow pelvis and avoid excessive use of linear stapler firings. In this study, we built an MRI-based deep-learning model to predict ≥3 linear stapler firings in LAR using DST anastomosis. This model aimed to help determine the surgical strategy for mid–low rectal cancer patients by predicting the probability of ≥3 firings of the linear stapler before surgery. Thus, we can reduce the occurrence of AL by using other more suitable anastomosis techniques. Our findings suggest that clinical information alone may not be sufficient to predict cases with ≥3 firings of the linear stapler. Compared with the clinical or image model, the integrated model that combined clinical information with pelvic MR images achieved better AUC and higher PPV. Akiyoshi et al. used clinical data and pelvic parameters to predict surgical difficulty and the incidence of AL in patients undergoing LAR using DST anastomosis. They found that tumor distance from the anal verge, BMI, pelvic outlet, and tumor invasive depth were independent predictors of operation time and occurrence of AL [10] . Foo et al. reported a predictive model for predicting ≥3 linear stapler firings. The model included the following parameters: tumor distance from the anal verge, gender, pelvic entrance, internodal distance and interspinous distance [40] . Compared with the above two studies, our predictive model has several advantages. (1) Using AI-based image segmentation, pelvic measurements can be identified comprehensively, rather than obtaining certain pelvic parameters separately. Thus, all anatomical features of the pelvis can be entirely integrated into the image model. (2) Clinically, the space between the pelvis, mesorectum and tumor mass affects the number of linear stapler firings. Our model takes into account not only pelvic parameters, but also the influence of meso-rectal factors and tumor conditions. (3) The predicting time of this AI-based warning model is only 100 ms. It greatly reduces the time and labor of manual measurement. It should be noted that our study has some limitations. First, the cohort of 328 patients was too small for training the deep-learning model, further study with larger sample size is needed. Second, other technical factors that were difficult to quantify can also affect the number of linear stapler firings, such as the correct angle between the stapler and rectum and precompression before stapler firings [35, 36] . Therefore, no 100% accuracy were achieved in our three models. Third, the number of linear stapler firings was just one of anastomotic factors related to AL, the circular end-to-end anastomosis, intersections of staple lines [41] , and the distance between the linear staple line [36] were also risk factors for AL. Finally, our deep-learning traning is only performed on T2-weighted MR sequences. Other MR sequences or contrast-enhanced MRI could be investigated in future studies. In conclusion, the pelvic MR-based deep-learning model can help identify the high-risk population with ≥3 linear stapler firings in laparoscopic LAR surgery. It might help determine the anastomotic technique for mid–low rectal cancer patients preoperatively. However, it is still necessary to verify its value through clinical application. Declarations The authors declare no conflict of interest. The datasets used during the current study available from the corresponding author on reasonable request. This study was funded by Shanghai Jiaotong University [grant number YG2019QNB24 to J.M.]. References Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6):394-424. https://doi.org/10.3322/caac.21492 MacFarlane J, Ryall R, Heald R (1993) Mesorectal excision for rectal cancer. Lancet (London, England) 341(8843):457-60. https://doi.org/10.1016/0140-6736(93)90207-w Rullier E, Rouanet P, Tuech J, Valverde A, Lelong B, Rivoire M, et al. (2017) Organ preservation for rectal cancer (GRECCAR 2): a prospective, randomised, open-label, multicentre, phase 3 trial. Lancet (London, England) 390(10093):469-79. https://doi.org/10.1016/s0140-6736(17)31056-5 Griffen F, Knight C, Whitaker J, Knight C (1990) The double stapling technique for low anterior resection. Results, modifications, and observations. Annals of surgery 211(6):745-51; discussion 51-2. https://doi.org/10.1097/00000658-199006000-00014 Shrikhande SV, Saoji RR, Barreto SG, Kakade AC, Waterford SD, Ahire SB, et al. (2007) Outcomes of resection for rectal cancer in India: The impact of the double stapling technique. World Journal of Surgical Oncology 5(1):35. https://doi.org/10.1186/1477-7819-5-35 Climent-Agustín M, Martin ST (2018) Complications of laparoscopic rectal cancer surgery. Mini-invasive Surgery 2:45. https://doi.org/10.20517/2574-1225.2018.62 Koedam T, Bootsma B, Deijen C, van de Brug T, Kazemier G, Cuesta M, et al. (2020) Oncological Outcomes After Anastomotic Leakage After Surgery for Colon or Rectal Cancer: Increased Risk of Local Recurrence. Annals of surgery 275(2):e420-e427. https://doi.org/10.1097/sla.0000000000003889 Kverneng Hultberg D, Svensson J, Jutesten H, Rutegård J, Matthiessen P, Lydrup M, et al. (2020) The Impact of Anastomotic Leakage on Long-term Function After Anterior Resection for Rectal Cancer. Diseases of the colon and rectum 63(5):619-28. https://doi.org/10.1097/dcr.0000000000001613 Taflampas P, Christodoulakis M, Tsiftsis D (2009) Anastomotic leakage after low anterior resection for rectal cancer: facts, obscurity, and fiction. Surgery today 39(3):183-8. https://doi.org/10.1007/s00595-008-3835-2 Akiyoshi T, Kuroyanagi H, Oya M, Konishi T, Fukuda M, Fujimoto Y, et al. (2009) Factors affecting the difficulty of laparoscopic total mesorectal excision with double stapling technique anastomosis for low rectal cancer. Surgery 146(3):483-9. https://doi.org/10.1016/j.surg.2009.03.030 Huang BY, Yang LC, Ding-Li M (2003) Modified rectocolon end-to-side anastomosis for middle and low rectal cancer after anterior resection. Journal of Modern Oncology 11(2):118-119. Ishii Y, Hasegawa H, Nishibori H, Endo T, Kitajima M (2006) The application of a new stapling device for open surgery (Contour Curved Cutter Stapler) in the laparoscopic resection of rectal cancer. Surgical endoscopy 20(8):1329-31. https://doi.org/10.1007/s00464-005-0633-4 Shearer R, Gale M, Aly OE, Aly EH (2013) Have early postoperative complications from laparoscopic rectal cancer surgery improved over the past 20 years? Colorectal Disease the Official Journal of the Association of Coloproctology of Great Britain & Ireland 15(10):1211-26. https://doi.org/10.1111/codi.12302. Bernhard, Dauser, Tamara, Braunschmid, Shahbaz, Ghaffari, et al. (2013) Anastomotic leakage after low anterior resection for rectal cancer: comparison of stapled versus compression anastomosis. Langenbeck's Archives of Surgery 398(7):957-64. https://doi.org/10.1007/s00423-013-1103-4 Kawada K, Sakai Y (2016) Preoperative, intraoperative and postoperative risk factors for anastomotic leakage after laparoscopic low anterior resection with double stapling technique anastomosis. World J Gastroenterol 22(25):5718-27. https://doi.org/10.3748/wjg.v22.i25.5718 Balciscueta Z, Uribe N, Caubet L, López M, Torrijo I, Tabet J, et al. (2020) Impact of the number of stapler firings on anastomotic leakage in laparoscopic rectal surgery: a systematic review and meta-analysis. Techniques in coloproctology 24(9):919-25. https://doi.org/10.1007/s10151-020-02240-7 Park JS, Choi GS, Kim SH, Kim HR, Kim NK, Lee KY, et al. (2013) Multicenter analysis of risk factors for anastomotic leakage after laparoscopic rectal cancer excision: the Korean laparoscopic colorectal surgery study group. Ann Surg 257(4):665-71. https://doi.org/10.1097/SLA.0b013e31827b8ed9 Ito M, Sugito M, Kobayashi A, Nishizawa Y, Tsunoda Y, Saito N (2008) Relationship between multiple numbers of stapler firings during rectal division and anastomotic leakage after laparoscopic rectal resection. Int J Colorectal Dis 23(7):703-7. https://doi.org/10.1007/s00384-008-0470-8 Zhang ZT (2019) Chinese Expert Consensus Statement on the Diagnostic, Prevention and Treation of the Anastomotic Leakage for Rectal Cancer (article in Chinese). Zhonghua Wei Chang Wai Ke Za Zhi 22(3):201-206. https://doi.org/10.3760/cma.j.issn.1671-0274.2019.03.001 Bertani E, Chiappa A, Della Vigna P, Radice D, Papis D, Cossu L, et al. (2014) The Impact of pelvimetry on anastomotic leakage in a consecutive series of open, laparoscopic and robotic low anterior resections with total mesorectal excision for rectal cancer. Hepatogastroenterology 61(134):1574-81. https://doi.org/10.5754/hge13724 Killeen T, Banerjee S, Vijay V, Al-Dabbagh Z, Francis D, Warren S (2010) Magnetic resonance (MR) pelvimetry as a predictor of difficulty in laparoscopic operations for rectal cancer. Surgical endoscopy 24(12):2974-9. https://doi.org/10.1007/s00464-010-1075-1 Zhou X, Su M, Hu K, Su Y, Ye Y, Huang C, et al. (2016) CT pelvimetry and clinicopathological parameters in evaluation of the technical difficulties in performing open rectal surgery for mid-low rectal cancer. Oncology letters 11(1):31-8. https://doi.org/10.3892/ol.2015.3827 Balyasnikova S, Brown G (2016) Optimal Imaging Strategies for Rectal Cancer Staging and Ongoing Management. Curr Treat Options Oncol 17(6):32. https://doi.org/10.1007/s11864-016-0403-7 Jhaveri KS, Sadaf A (2009) Role of MRI for staging of rectal cancer. Expert Rev Anticancer Ther 9(4):469-81. https://doi.org/10.1586/era.09.13 Rajjak S, Kureshi AK (2021) Multiple-Object Detection and Segmentation Based on Deep Learning in High-Resolution Video Using Mask-RCNN. International Journal of Pattern Recognition and Artificial Intelligence 35(13). https://doi.org/10.1142/S0218001421500385 Tran D, Bourdev L, Fergus R, Torresani L, Paluri M (2015) Learning Spatiotemporal Features with 3D Convolutional Networks. IEEE International Conference on Computer Vision. pp, 4489-4497. https://doi.org/10.1109/ICCV.2015.510 Li W, Wang G, Fidon L, Ourselin S, Cardoso MJ, Vercauteren T (2017) On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task. Springer, Cham 10265: 348–360. https://doi.org/10.1007/978-3-319-59050-9_28 Meng J, Xue L, Chang Y, Zhang J, Yang K (2020) Automatic detection and segmentation of adenomatous colorectal polyps during colonoscopy using Mask R-CNN. Open Life Sciences 15(1):588-96. https://doi.org/10.1515/biol-2020-0055 Qadir HA, Shin Y, Solhusvik J, Bergsland J, Balasingham I (2019) Polyp Detection and Segmentation using Mask R-CNN: Does a Deeper Feature Extractor CNN Always Perform Better? 2019 13th International Symposium on Medical Information and Communication Technology (ISMICT) pp, 1-6. https://doi.org/10.1109/ISMICT.2019.8743694 Russell BC, Torralba A, Murphy KP, Freeman WT (2008) LabelMe: A Database and Web-Based Tool for Image Annotation. International Journal of Computer Vision 77(1-3) : 157–173. https://doi.org/10.1007/s11263-007-0090-8 Lin TY, Maire M, Belongie S, Hays J, Zitnick CL (2014) Microsoft COCO: Common Objects in Context. Springer International Publishing. https://doi.org/10.1007/978-3-319-10602-1_48 Hu J (2013) Proceedings of the 2013 IEEE International Conference on Computer Vision. IEEE International Conference on Computer Vision. Huang MJ, Liang JL, Wang H, Kang L, Deng YH, Wang JP (2011) Laparoscopic-assisted versus open surgery for rectal cancer: a meta-analysis of randomized controlled trials on oncologic adequacy of resection and long-term oncologic outcomes. International Journal of Colorectal Disease 26(4):415. https://doi.org/10.1007/s00384-010-1091-6 Kim C, Baek S, Hur H, Min B, Baik S, Kim N (2016) Anastomotic Leakage After Low Anterior Resection for Rectal Cancer Is Different Between Minimally Invasive Surgery and Open Surgery. Annals of surgery 263(1):130-7. https://doi.org/10.1097/sla.0000000000001157 Kawada K, Hasegawa S, Hida K, Hirai K, Okoshi K, Nomura A, et al. (2014) Risk factors for anastomotic leakage after laparoscopic low anterior resection with DST anastomosis. Surgical Endoscopy 28(10):2988-95. https://doi.org/10.1007/s00464-014-3564-0 Kuroyanagi H, Oya M, Ueno M, Fujimoto Y, Yamaguchi T, Muto T (2008) Standardized technique of laparoscopic intracorporeal rectal transection and anastomosis for low anterior resection. Surgical endoscopy 22(2):557-61. https://doi.org/10.1007/s00464-007-9626-9 Nakagoe T, Ishikawa H, Sawai T, Tsuji T, Takeshita H, Nanashima A, et al. (2005) Oncological outcome of ultra-low anterior resection with total mesorectal excision for carcinoma of the lower third of the rectum: Comparison of intrapelvic double-stapled anastomosis and transanal coloanal anastomosis. Hepato-gastroenterology 52(66):1692-7. https://doi.org/10.1246/bcsj.40.2063 Spinelli A, Foppa C, Carvello M, Sacchi M, De Lucia F, Clerico G, et al. (2021) Transanal Transection and Single-Stapled Anastomosis (TTSS): A comparison of anastomotic leak rates with the double-stapled technique and with transanal total mesorectal excision (TaTME) for rectal cancer. European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology 47(12):3123-9. https://doi.org/10.1016/j.ejso.2021.08.002 Penna M, Hompes R, Arnold S, Wynn G, Austin R, Warusavitarne J, et al. (2019) Incidence and Risk Factors for Anastomotic Failure in 1594 Patients Treated by Transanal Total Mesorectal Excision: Results From the International TaTME Registry. Annals of surgery 269(4):700-11. https://doi.org/10.1097/sla.0000000000002653 Foo C, Hung H, Ho Y, Lam W, Law W (2020) Predicting the level of difficulty of the double-stapling technique in laparoscopic total mesorectal excision. Surgical endoscopy 34(8):3382-7. https://doi.org/10.1007/s00464-019-07112-2 Lee S, Ahn B, Lee S (2017) The Relationship Between the Number of Intersections of Staple Lines and Anastomotic Leakage After the Use of a Double Stapling Technique in Laparoscopic Colorectal Surgery. Surgical laparoscopy, endoscopy & percutaneous techniques 27(4):273-81. https://doi.org/10.1097/sle.0000000000000422 Tables Table 1 Clinical characteristics of patients in deep-learning set Number of linear stapler firings ≥3 ≤2 n = 58 (17.7%) n = 270 (82.3%) p value Baseline characteristics Age [y] ＞70 ≤70 16 (27.6) 42 (72.4) 54 (20.0) 216 (80.0) 0.218 Gender, n (%) 0.084 Male 46 (79.3) 181 (67.0) Female 12 (20.7) 89 (33.0) BMI [Kg/m2] ＞25 ≤25 17 (29.3) 41 (70.7) 72 (26.7) 198 (73.3) 0.745 Diabetes mellitus, n (%) 0.873 Yes 9 (15.5) 32 (11.9) No 49 (84.5) 238 (88.1) nCRT, n (%) 0.709 Yes 14 (24.1) 71 (26.3) No 44 (75.9) 199 (73.7) Operation time [min] >150 ≤150 32 (55.2) 26 (44.8) 124 (45.9) 146 (54.1) 0.246 Anastomotic leakage, n (%) 0.021 Yes 14 (24.1) 34 (12.6) No 44 (75.9) 236 (87.4) Biochemical data Hemoglobin [g/L] <120 ≥120 12 (20.7) 46 (79.3) 58 (21.5) 212 (78.5) 0.893 Albumin [g/L] 5 26 (44.8) 71 (26.3) 0.007 ≤5 32 (55.2) 199 (73.7) Tumor characteristics Distance from anus [cm] <5 ≥5 4 (6.9) 54 (93.1) 39 (14.4) 231 (85.6) 0.138 Tumor size [cm] ≥5 <5 14 (24.1) 44 (75.8) 26 (9.6) 244 (90.4) 0.004 T stage, n (%) 0.865 T≤2 14 (24.1) 68 (25.2) T3-4 44 (75.9) 229 (84.8) N stage, n (%) 0.663 N0 30(51.7) 152 (56.3) N+ 28(48.3) 118 (43.7) CRM (evaluated by MRI), n (%) 0.014 Positive 20 (34.5) 52 (19.3) Negative 38 (65.5) 218 (80.7) Abbreviation: MRI: magnetic resonance imaging; CRM: circumferential resection margin; N: node; T: tumor; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index. Table 2 Risk factors of ≥3 linear stapler firings in deep-learning set Factors Univariate analysis Multivariate analysis OR (95% CI) p value OR (95% CI) p value Age [y] (≥70/<70) 1.45 (0.82, 2.56) 0.202 N.A. N.A. Gender (Male/Female) 1.89 (0.95, 3.74) 0.069 1.82 (0.90, 3.71) 0.098 BMI [Kg/m2] (≥25/<25) 1.14 (0.61, 2.13) 0.681 N.A. N.A. Diabetes mellitus (Y/N) 1.37 (0.61, 3.04) 0.445 N.A. N.A. nCRT (Y/N) 0.89 (0.46, 1.73) 0.734 N.A. N.A. Albumin [g/L] (5/≤5) 2.39 (1.33, 4.32) 0.004 2.20 (1.20, 4.04) 0.011 Distance from anus [cm] (<5/≥5) 2.28 (0.78, 6.65) 0.132 N.A. N.A. Tumor size [cm] (≥5/<5) 2.99 (1.45, 6.16) 0.003 2.54 (1.15, 5.61) 0.021 CRM (evaluated by MRI) (+/-) 2.21 (1.19, 4.10) 0.012 1.66 (0.84, 3.25) 0.143 Abbreviation: MRI: magnetic resonance imaging; CRM: circumferential resection margin; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index; N.A.: not applicable; OR: odds ratio; CI: confidence interval. Table 3 Comparison of clinical characteristics between deep-learning set and validation set Deep-learning set Validation set p value n = 328 n=128 Baseline characteristics Age [y] ＞70 ≤70 70 (21.3) 258 (78.0) 36 (28.1) 92 (71.9) 0.139 Gender, n (%) 0.561 Male 227 (69.2) 83 (64.8) Female 101 (30.8) 45 (35.2) BMI [Kg/m2] ＞25 ≤25 89 (27.1) 239 (72.9) 42 (32.8) 86 (67.2) 0.250 Diabetes mellitus, n (%) 0.342 Yes 41 (12.5) 23 (18.0) No 287 (87.5) 105 (82.0) nCRT, n (%) 0.201 Yes 85 (25.9) 41 (32.0) No 243 (74.1) 87 (68.0) Number of linear stapler cartridges, n (%) 0.204 ≥3 58 (17.7) 16 (12.5) ≤2 270 (82.3) 112 (87.5) Anastomotic leakage, n (%) 0.090 Yes 48 (14.6) 11 (8.6) No 280 (85.4) 117 (91.4) Biochemical data Albumin [g/L] 5 97 (29.6) 32 (25.0) ≤5 226 (68.9) 96 (75.0) Tumor characteristics Distance from anus [cm] <5 ≥5 43 (13.1) 285 (86.9) 14 (10.9) 114 (89.1) 0.637 Tumor size [cm], n (%) 0.130 ≥5 40 (12.2) 9 (7.0) <5 288 (87.8) 119 (93.0) CRM (evaluated by MRI), n (%) 0.092 Positive 72 (22.0) 19 (14.8) Negative 256 (78.0) 109 (69.5) Abbreviation: MRI: magnetic resonance imaging; CRM: circumferential resection margin; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 02 Nov, 2023 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Major revision 23 Sep, 2023 Reviews received at journal 21 Sep, 2023 Reviewers agreed at journal 11 Sep, 2023 Reviews received at journal 10 Apr, 2023 Reviewers agreed at journal 09 Apr, 2023 Reviewers agreed at journal 08 Apr, 2023 Reviewers invited by journal 08 Apr, 2023 Editor assigned by journal 27 Mar, 2023 Editor invited by journal 27 Mar, 2023 Submission checks completed at journal 27 Mar, 2023 First submitted to journal 11 Mar, 2023 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-2681419","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":186800751,"identity":"6ee72804-cf80-4b0f-bb68-651a7f9f452e","order_by":0,"name":"Zhanwei Fu","email":"","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhanwei","middleName":"","lastName":"Fu","suffix":""},{"id":186800752,"identity":"13775a4c-b2b6-4246-a054-bbd6d0277616","order_by":1,"name":"Shuchun Li","email":"","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Shuchun","middleName":"","lastName":"Li","suffix":""},{"id":186800753,"identity":"9ccc84f4-0c9f-41cd-86e3-19a26cf751e9","order_by":2,"name":"Lu Zang","email":"","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lu","middleName":"","lastName":"Zang","suffix":""},{"id":186800754,"identity":"4db4ca70-9b8e-4ab4-828a-924ff20e11d6","order_by":3,"name":"Feng Dong","email":"","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Feng","middleName":"","lastName":"Dong","suffix":""},{"id":186800755,"identity":"72984daa-286f-4c5d-b4c6-9202e05943ce","order_by":4,"name":"Zhenghao Cai","email":"","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhenghao","middleName":"","lastName":"Cai","suffix":""},{"id":186800756,"identity":"d99f66de-faa2-4694-a7e9-02fa6ba60e53","order_by":5,"name":"Junjun Ma","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYBADHn72/o8PEipqiNciI9lzwNjgwZljxGuxMbiRYCb5sIWZsFLd9rOPPxdUHOYBakmrSGxgY+Bv707Aq8XsTLqZ9Iwzh3kkzzw4diNxhwyDxJmzG/BrOZDGxszbdpuH73hi243EM2wMBhK5BLScf8b8mfffbR6GA8lsBYltzERouZHGIM3bcJtH4EQaGwORWp6xSfMc+88j2XOGWSLhzDEewn45n8b8macmzZ6fvYfx44+KGjn+9l78WjAAD2nKR8EoGAWjYBRgBQDHwkpuJ0upJAAAAABJRU5ErkJggg==","orcid":"","institution":"Ruijin Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Junjun","middleName":"","lastName":"Ma","suffix":""}],"badges":[],"createdAt":"2023-03-11 14:44:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2681419/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2681419/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-023-46225-6","type":"published","date":"2023-11-02T15:00:43+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":34984815,"identity":"3dc01232-8815-4c64-aec5-ea155b2e46fd","added_by":"auto","created_at":"2023-03-29 14:22:13","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":359125,"visible":true,"origin":"","legend":"\u003cp\u003eExamples of target regions. A, B: manually labeled; C, D: segmented by Mask region-based convolutional neural network (Mask R-CNN) based model (The regions of tumor body, mesorectum, and pelvis were represented by green, yellow, and drab, respectively)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-2681419/v1/261c62ea1381af7cd92c0b2d.png"},{"id":34984814,"identity":"c964b63f-7200-44b0-9f5e-f20a64aa9583","added_by":"auto","created_at":"2023-03-29 14:22:13","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":115800,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the design. BMI: Body mass index; CEA: Carcinoembryonic antigen; CRM: Circumferential resection margin; MR: Magnetic resonance; Mask R-CNN: Mask region-based convolutional neural network.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-2681419/v1/962d3649f013d1d8be81e664.png"},{"id":34984816,"identity":"f7fa99f2-73b5-4140-b75a-acdd373551d3","added_by":"auto","created_at":"2023-03-29 14:22:13","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":75492,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver operating characteristic curves of the predictive models. A: clinical model in deep-learning set; B: image model in deep-learning set; C: integrated model in deep-learning set; D: clinical model in validation set; E: image model in validation set; F: integrated model in validation set\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-2681419/v1/0d37bbaafa4fab0cb8a9f54b.png"},{"id":45943158,"identity":"58d8bf86-693d-4728-8fdd-ebc8289767c2","added_by":"auto","created_at":"2023-11-06 15:08:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":818784,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2681419/v1/d8252ae5-3f19-4461-9969-2069a6c2bc90.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model","fulltext":[{"header":"Introduction","content":"\u003cp\u003eColorectal cancer (CRC) is the third most prevalent cancer worldwide with high rate of cancer mortality, while approximately one third of all CRCs occur in rectum\u003csup\u003e[1]\u003c/sup\u003e. Total mesorectal excision (TME) is the primary treatment for localized rectal cancer patients, and has high rate of cancer control\u0026nbsp;\u003csup\u003e[2, 3]\u003c/sup\u003e. It is a widely applied surgical technique to perform laparoscopic low anterior resection (LAR) using double-stapling technique (DST) anastomosis\u0026nbsp;\u003csup\u003e[4, 5]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn mid\u0026ndash;low rectal cancer patients, anastomotic leakage (AL) is the most common postoperative complication after LAR\u0026nbsp;\u003csup\u003e[6]\u003c/sup\u003e. In addition to increasing the incidence of reoperation and mortality, AL may even negatively affect long-term survival\u0026nbsp;\u003csup\u003e[7-9]\u003c/sup\u003e. Despite advances in surgical techniques and anastomotic equipment over the past few decades\u003csup\u003e[10-12]\u003c/sup\u003e, there has been no significant decrease in the incidence of AL after LAR\u0026nbsp;\u003csup\u003e[13]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eDST simplifies colorectal reconstruction in LAR, especially in laparoscopic surgery, but the incidence of postoperative AL was not reduced by the application of this technique\u003csup\u003e[14, 15]\u003c/sup\u003e. Several studies have proved that \u0026ge;3 linear stapler firings in laparoscopic LAR is an independent risk factor for AL\u0026nbsp;\u003csup\u003e[16-18]\u003c/sup\u003e. Thus, the Chinese expert consensus statement proposes \u0026le;2 linear stapler firings in LAR surgery\u003csup\u003e[19]\u003c/sup\u003e. To avoid multiple linear stapler firings, a predictive model should be created to predict\u0026nbsp;\u0026ge;3 linear stapler firings in anastomosis, and alternative techniques could be considered.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSeveral studies have shown that pelvic anatomical features (such as the anteroposterior diameter and the transverse diameter of the pelvic outlet, the anteroposterior diameter of the pelvic inlet) are related to surgical difficulty and the number of the linear stapler firings\u0026nbsp;\u003csup\u003e[20-22]\u003c/sup\u003e. Those studies mainly measured pelvic parameters, but did not include the influence of rectal and mesenteric conditions. How to effectively integrate above parameters to predict\u0026nbsp;\u0026ge;3 linear stapler firing needs further research.\u003c/p\u003e\n\u003cp\u003eWith the advancement of imaging technology, pelvic magnetic resonance imaging (MRI) is a preferred tool for local staging of mid-low rectal cancer before surgery\u003csup\u003e[23, 24]\u003c/sup\u003e. MRI can obtain relevant parameters of tumor, meso-rectum and pelvis comprehensively and accurately. Mask region-based convolutional neural network (Mask R-CNN)\u0026nbsp;\u003csup\u003e[25]\u003c/sup\u003e and three-dimensional convolutional network (C3D)\u0026nbsp;\u003csup\u003e[26]\u003c/sup\u003e image recognition are current techniques for advanced-recognition artificial intelligence (AI), and have been applied in various medical fields\u0026nbsp;\u003csup\u003e[27-29]\u003c/sup\u003e. With advanced deep-learning technology, pelvic MRIs\u0026rsquo; complex data can be identified, extracted, analyzed, and integrated efficiently, and we can create a predictive model to screen out high-risk patients with \u0026ge;3 linear stapler firings based on the database.\u003c/p\u003e\n\u003cp\u003eIn this study, we aimed to establish the risk factors for \u0026ge;3 linear stapler firings in laparoscopic LAR using DST anastomosis, and to create and validate a predictive model for \u0026ge;3 linear stapler firings with deep-learning technology based on MRI.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003ePatients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA total of 328 mid\u0026ndash;low rectal cancer patients who underwent laparoscopic LAR at Ruijin Hospital Affiliated to Shanghai Jiaotong University School of Medicine, China, between January 2016 and June 2021 were retrospectively analyzed as the deep-learning set. Clinical data were obtained from Ruijin hospital database and medical records. All methods were performed in accordance with the relevant guidelines and regulations, and the study was approved by the Medical Ethics Committee of Ruijin Hospital (No. 2019-82). The need for informed consent was waived by Ethics Committee of Ruijin Hospital. With a split ratio of 4:1, 260 patients were divided into the training set and 68 patients were divided into the testing set on the basis of an unbiased random sampling method. The external validation set comprised 128 patients from an independent clinical trial in our institution (Transanal versus laparoscopic total mesorectal excision for rectal cancer, ClinicalTrials.gov Identifier: NCT03359616).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe inclusion criteria were: (1) histopathologically confirmed rectal cancer; (2) pelvic MRI examination within 14 days before surgery; (3) tumor distance from the anal verge \u0026le;10 cm; and (4) laparoscopic LAR with DST anastomosis. The exclusion criteria were: (1) LAR without anastomosis (e.g., Hartmann\u0026rsquo;s operation); (2) anastomotic techniques rather than DST (e.g., manual rectal anastomosis through the anus); (3) number of linear stapler firings was not recorded on surgical reports; and (4)\u0026nbsp;robotic surgery.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurgical procedure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLaparoscopic LAR was performed by the gastrointestinal surgery team with experience in completing more than 100 rectal cancer operations every year. The laparoscopic LAR surgical procedure was carried out in strict accordance with the national guidelines for laparoscopic radical resection of CRC (2018 edition). During dissection of the distal rectum, the surgeon manually fired endoscopic linear staplers (Endo-GIA\u003csup\u003eTM\u003c/sup\u003e Ultra Universal Stapler Reload with Tri-stapler\u003csup\u003eTM\u003c/sup\u003e Technology; Covidien LLC), which was loaded with 60- or 45-mm staple cartridges which have three types of heights: 3.0, 3.5 and 4.0 mm.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData collection and model building\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe collected clinical data that potentially correlated with the number of linear staplers used in surgery, including baseline characteristics, including age, gender and body mass index (BMI); biochemical data, including hemoglobin, albumin, and carcinoma embryonic antigen (CEA); and tumor characteristics, including tumor distance from the anal verge, tumor stage, tumor size and circumferential resection margin (CRM). A predictive model of \u0026ge;3 linear stapler firings was constructed by binary logistic regression. The variables of the clinical model included: gender, BMI, serum CEA level (\u0026gt; or \u0026le;5 ng/mL), tumor distance from the anal verge, tumor size and CRM.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI and target area labeling\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring MRI, patients were in the supine position and scanned using a Philips INGENIA\u003csup\u003eTM\u003c/sup\u003e scanner with 3.0 T field strength. The pelvic phased-array surface coil covered from the aortic bifurcation to the anal verge. The scanning parameters were: layer thickness 5 mm; field of view 250\u0026acute;340\u0026acute;166 mm; echo time 80 ms; repetition time 3565 ms; and image matrix 312\u0026acute;357.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWith the Picture Archiving and Communication System, fat-suppressed fast spin-echo (FSE) T2-weighted sequences in the axial plane of the pelvis were used for image segmentation. Then, pelvic MRI specialists who have \u0026gt;10 years of experience built an image database by an online annotation tool called Labelme (labelme.csail.mit.edu/)\u0026nbsp;\u003csup\u003e[30]\u003c/sup\u003e, and labeled three kinds of target area on each of the T2-weighted images (tumor body, mesorectum, and pelvis represented by green, yellow, and drab, respectively, Figure 1A-B). Then, all data were transformed into a COCO dataset for segmentation experiments\u0026nbsp;\u003csup\u003e[31]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSegmentation model based on Mask R-CNN\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs an effective two-stage detection and segmentation algorithm, Mask R-CNN was adopted to identify and segment the three kinds of target areas in the image (Figure 1C-D). In the first stage, the ResNet-101-FPN network served as the backbone to extract multiscale and discriminative feature maps. The Region Proposal Network scanned the feature map in a sliding-window and selected the rough detection rectangle that contained the object. After the regions of interest alignment process, the candidate regions entered the next stage. This consisted of three functional branches: classification, detection and segmentation, based on fully connected layers and convolutional layers. We trained the Mask R-CNN network on the training set for 200 epochs, and evaluated the performance of the testing set with standard COCO metrics. We evaluated the trained Mask R-CNN model, obtained average precision (AP) by calculating the precision\u0026ndash;recall curve under different intersection-over-union thresholds, and then calculated the three types of targets. The respective AP values of the regions were weighted to obtain the class-wide mean average precision (mAP). mAP\u0026gt;50 indicated that the model performed well\u0026nbsp;\u003csup\u003e[32]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeep-learning model based on C3D\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used C3D networks to address 24 images of a case simultaneously and learn 3D spatial features. The C3D network consisted of eight 3D-convolution layers, a softmax layer, two fully connected layers and five pooling layers. It took the entire image of the case as input and output the probability of \u0026ge;3 linear stapler firings, and the sample with probability \u0026gt;50% (empirical value) was judged as positive. The C3D network was trained until convergence (~1000 epochs) and evaluated the performance of the deep-learning model. In the training set, two C3D-based models were trained, including an image model using only MR images and an integrated model based on both MR images and the six clinical variables in clinical model. In the testing set, the clinical model, image model and integrated model were examined, and receiver operating characteristic (ROC) curves were plotted. Area under the curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV), and Youden index were calculated.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExternal validation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe used clinical data and T2-weighted images of patients from the validation set to externally validate the predictive performance of the above three models. ROC curves were plotted, and sensitivity, specificity, accuracy, Youden index, PPV, and AUC were calculated. The flow chart of the design is shown in Figure 2.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStatistical analysis was performed using SPSS (version 25.0). Categorical variables were analyzed by Fisher\u0026rsquo;s exact test or Pearson\u0026rsquo;s chi-square. In our binary logistic regression models, only factors with a P value\u0026lt;0.10 in the univariate analysis were entered into the multivariate analysis. All tests were two-sided, and differences were considered statistically significant at p\u0026lt;0.05.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eClinical characteristics of patients in deep-learning set\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe 328 patients had a median age of 63 (24\u0026ndash;87) years, including 227 men and 101 women. The proportion of patients who received \u0026ge;3 linear stapler firings was 17.7% (58/328). Clinical characteristics of patients with \u0026ge;3 firings of the linear stapler and those with \u0026le;2 firings in the deep-learning set were compared (Table 1). There were no significant differences in age, gender, BMI, diabetes mellitus, neoadjuvant chemoradiotherapy ratio, operation time, hemoglobin, albumin, tumor distance from the anal verge, T stage and N stage between the two groups. Patients with \u0026ge;3 firings showed significantly higher incidence of AL than those with \u0026le;2 firings (p=0.021), and patients with \u0026ge;3 firings of the linear stapler showed significantly higher CEA level (p=0.007), larger tumor size (p=0.004) and higher rate of positive CRM (p=0.014). In univariate and multivariate analyses, tumor size \u0026ge;5 cm (odds ratio (OR)=2.54, 95% confidence interval (CI)=1.15\u0026ndash;5.60, p=0.021) and serum CEA \u0026gt;5 ng/mL [OR=2.20, 95% CI=1.20\u0026ndash;4.04, p=0.011] were independent risk factors associated with \u0026ge;3 linear stapler firings (Table 2).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePredicting performance in testing set\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe AUCs of the clinical, imaging and integrated models were obtained as 0.72, 0.81 and 0.88, respectively (Figure 3A\u0026ndash;C). The sensitivity, specificity, accuracy, PPV and Youden index of the clinical model were: 70.0%, 81.0%, 79.4%, 38.9%, and 0.51, respectively. The relevant indicators of the image model were: 50.0%, 98.3%, 91.2%, 83.3%, and 0.48, respectively. The relevant indicators of the integrated model were: 70.0%, 98.3%, 94.1%, 87.5%, and 0.68, respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExternal validation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the validation set, the proportion of patients who received\u0026nbsp;\u0026ge;3 linear stapler firings was 12.5% (16/128). All related clinical characteristics between the deep-learning and validation sets were comparable (Table 3). The AUCs of the clinical, imaging and integrated models were obtained as 0.65, 0.75 and 0.84, respectively (Figure 3D\u0026ndash;F).\u003c/p\u003e\n\u003cp\u003eThe sensitivity, specificity, accuracy, PPV and Youden index of the clinical model were: 62.5%, 66.1%, 65.6%, 21.0%, and 0.29, respectively. The relevant indicators of the image model were: 68.8%, 95.5%, 92.1%, 68.8%, and 0.64, respectively. The relevant indicators of the integrated model were: 68.8%, 97.3%, 93.8%, 78.5%, and 0.66, respectively.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eLAR using DST anastomosis is currently a widely applied surgical technique for mid–low rectal cancer, a series of high-quality randomized controlled trials has confirmed its feasibility and safety\u0026nbsp;\u003csup\u003e[5, 33]\u003c/sup\u003e. The technique greatly reduces the difficulty of reconstruction of the digestive tract. However, some studies have reported that multiple linear stapler firings is closely related to AL\u0026nbsp;\u003csup\u003e[17, 34]\u003c/sup\u003e, and AL is more likely to occur at the intersection of two staples\u0026nbsp;\u003csup\u003e[35]\u003c/sup\u003e. In some cases, due to the limitation of the pelvic space or thickness of the rectum, the surgeons have to trigger more linear stapler firings during rectal dissection\u0026nbsp;\u003csup\u003e[36]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn the high-risk populations with ≥3 linear stapler firings, other anastomosis techniques rather than DST anastomosis could be considered, such as transanal anastomosis after transanal transection of the rectum\u0026nbsp;\u003csup\u003e[37]\u003c/sup\u003e and manual purse-string suture after endoluminal transection of the rectum (e.g., Transanal total mesorectal excision)\u0026nbsp;\u003csup\u003e[38]\u003c/sup\u003e. Although some studies have shown that the above techniques do not reduce the incidence of AL\u0026nbsp;\u003csup\u003e[39]\u003c/sup\u003e, these techniques can minimize the anastomotic difficulty in patients with a narrow pelvis and avoid excessive use of linear stapler firings.\u003c/p\u003e\n\u003cp\u003eIn this study, we built an MRI-based deep-learning model to predict\u0026nbsp;≥3 linear stapler firings in LAR using DST anastomosis. This model aimed to help determine the surgical strategy for mid–low rectal cancer patients by predicting the probability of ≥3 firings of the linear stapler before surgery. Thus, we can reduce the occurrence of AL by using other more suitable anastomosis techniques. Our findings suggest that clinical information alone may not be sufficient to predict cases with ≥3 firings of the linear stapler. Compared with the clinical or image model, the integrated model that combined clinical information with pelvic MR images achieved better AUC and higher PPV.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAkiyoshi et al. used clinical data and pelvic parameters to predict surgical difficulty and the incidence of AL in patients undergoing LAR using DST anastomosis. They found that tumor distance from the anal verge, BMI, pelvic outlet, and tumor invasive depth were independent predictors of operation time and occurrence of AL\u0026nbsp;\u003csup\u003e[10]\u003c/sup\u003e. Foo et al. reported a predictive model for predicting ≥3 linear stapler firings. The model included the following parameters: tumor distance from the anal verge, gender, pelvic entrance, internodal distance and interspinous distance\u0026nbsp;\u003csup\u003e[40]\u003c/sup\u003e. Compared with the above two studies, our predictive model has several advantages. (1) Using AI-based image segmentation, pelvic measurements can be identified comprehensively, rather than obtaining certain pelvic parameters separately. Thus, all anatomical features of the pelvis can be entirely integrated into the image model. (2) Clinically, the space between the pelvis, mesorectum and tumor mass affects the number of linear stapler firings. Our model takes into account not only pelvic parameters, but also the influence of meso-rectal factors and tumor conditions. (3) The predicting time of this AI-based warning model is only 100 ms. It greatly reduces the time and labor of manual measurement.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt should be noted that our study has some limitations. First, the cohort of 328 patients was too small for training the deep-learning model, further study with larger sample size is needed. Second, other technical factors that were difficult to quantify can also affect the number of linear stapler firings, such as the correct angle between the stapler and rectum and precompression before stapler firings\u0026nbsp;\u003csup\u003e[35, 36]\u003c/sup\u003e. Therefore, no 100% accuracy were achieved in our three models. Third, the number of linear stapler firings was just one of anastomotic factors related to AL, the circular end-to-end anastomosis, intersections of staple lines\u0026nbsp;\u003csup\u003e[41]\u003c/sup\u003e, and the distance between the linear staple line\u0026nbsp;\u003csup\u003e[36]\u003c/sup\u003e were also risk factors for AL. Finally, our deep-learning traning is only performed on T2-weighted MR sequences. Other MR sequences or contrast-enhanced MRI could be investigated in future studies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn conclusion, the pelvic MR-based deep-learning model can help identify the high-risk population with ≥3 linear stapler firings in laparoscopic LAR surgery. It might help determine the anastomotic technique for mid–low rectal cancer patients preoperatively. However, it is still necessary to verify its value through clinical application.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003eThe datasets used during the current study available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003eThis study was funded by Shanghai Jiaotong University [grant number YG2019QNB24 to J.M.].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eBray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68(6):394-424. https://doi.org/10.3322/caac.21492\u003c/li\u003e\n \u003cli\u003eMacFarlane J, Ryall R, Heald R (1993) Mesorectal excision for rectal cancer. Lancet (London, England) 341(8843):457-60. https://doi.org/10.1016/0140-6736(93)90207-w\u003c/li\u003e\n \u003cli\u003eRullier E, Rouanet P, Tuech J, Valverde A, Lelong B, Rivoire M, et al. (2017) Organ preservation for rectal cancer (GRECCAR 2): a prospective, randomised, open-label, multicentre, phase 3 trial. Lancet (London, England) 390(10093):469-79. https://doi.org/10.1016/s0140-6736(17)31056-5\u003c/li\u003e\n \u003cli\u003eGriffen F, Knight C, Whitaker J, Knight C (1990) The double stapling technique for low anterior resection. Results, modifications, and observations. Annals of surgery 211(6):745-51; discussion 51-2. https://doi.org/10.1097/00000658-199006000-00014\u003c/li\u003e\n \u003cli\u003eShrikhande SV, Saoji RR, Barreto SG, Kakade AC, Waterford SD, Ahire SB, et al. (2007) Outcomes of resection for rectal cancer in India: The impact of the double stapling technique. World Journal of Surgical Oncology 5(1):35. https://doi.org/10.1186/1477-7819-5-35\u003c/li\u003e\n \u003cli\u003eCliment-Agust\u0026iacute;n M, Martin ST (2018) Complications of laparoscopic rectal cancer surgery. Mini-invasive Surgery 2:45. https://doi.org/10.20517/2574-1225.2018.62\u003c/li\u003e\n \u003cli\u003eKoedam T, Bootsma B, Deijen C, van de Brug T, Kazemier G, Cuesta M, et al. (2020) Oncological Outcomes After Anastomotic Leakage After Surgery for Colon or Rectal Cancer: Increased Risk of Local Recurrence. Annals of surgery 275(2):e420-e427. https://doi.org/10.1097/sla.0000000000003889\u003c/li\u003e\n \u003cli\u003eKverneng Hultberg D, Svensson J, Jutesten H, Ruteg\u0026aring;rd J, Matthiessen P, Lydrup M, et al. (2020) The Impact of Anastomotic Leakage on Long-term Function After Anterior Resection for Rectal Cancer. Diseases of the colon and rectum 63(5):619-28. https://doi.org/10.1097/dcr.0000000000001613\u003c/li\u003e\n \u003cli\u003eTaflampas P, Christodoulakis M, Tsiftsis D (2009) Anastomotic leakage after low anterior resection for rectal cancer: facts, obscurity, and fiction. Surgery today 39(3):183-8. https://doi.org/10.1007/s00595-008-3835-2\u003c/li\u003e\n \u003cli\u003eAkiyoshi T, Kuroyanagi H, Oya M, Konishi T, Fukuda M, Fujimoto Y, et al. (2009) Factors affecting the difficulty of laparoscopic total mesorectal excision with double stapling technique anastomosis for low rectal cancer. Surgery 146(3):483-9. https://doi.org/10.1016/j.surg.2009.03.030\u003c/li\u003e\n \u003cli\u003eHuang BY, Yang LC, Ding-Li M (2003) Modified rectocolon end-to-side anastomosis for middle and low rectal cancer after anterior resection. Journal of Modern Oncology 11(2):118-119.\u003c/li\u003e\n \u003cli\u003eIshii Y, Hasegawa H, Nishibori H, Endo T, Kitajima M (2006) The application of a new stapling device for open surgery (Contour Curved Cutter Stapler) in the laparoscopic resection of rectal cancer. Surgical endoscopy 20(8):1329-31. https://doi.org/10.1007/s00464-005-0633-4\u003c/li\u003e\n \u003cli\u003eShearer R, Gale M, Aly OE, Aly EH (2013) Have early postoperative complications from laparoscopic rectal cancer surgery improved over the past 20 years? Colorectal Disease the Official Journal of the Association of Coloproctology of Great Britain \u0026amp; Ireland 15(10):1211-26. https://doi.org/10.1111/codi.12302.\u003c/li\u003e\n \u003cli\u003eBernhard, Dauser, Tamara, Braunschmid, Shahbaz, Ghaffari, et al. (2013) Anastomotic leakage after low anterior resection for rectal cancer: comparison of stapled versus compression anastomosis. Langenbeck\u0026apos;s Archives of Surgery 398(7):957-64. https://doi.org/10.1007/s00423-013-1103-4\u003c/li\u003e\n \u003cli\u003eKawada K, Sakai Y (2016) Preoperative, intraoperative and postoperative risk factors for anastomotic leakage after laparoscopic low anterior resection with double stapling technique anastomosis. World J Gastroenterol 22(25):5718-27. https://doi.org/10.3748/wjg.v22.i25.5718\u003c/li\u003e\n \u003cli\u003eBalciscueta Z, Uribe N, Caubet L, L\u0026oacute;pez M, Torrijo I, Tabet J, et al. (2020) Impact of the number of stapler firings on anastomotic leakage in laparoscopic rectal surgery: a systematic review and meta-analysis. Techniques in coloproctology 24(9):919-25. https://doi.org/10.1007/s10151-020-02240-7\u003c/li\u003e\n \u003cli\u003ePark JS, Choi GS, Kim SH, Kim HR, Kim NK, Lee KY, et al. (2013) Multicenter analysis of risk factors for anastomotic leakage after laparoscopic rectal cancer excision: the Korean laparoscopic colorectal surgery study group. Ann Surg 257(4):665-71. https://doi.org/10.1097/SLA.0b013e31827b8ed9\u003c/li\u003e\n \u003cli\u003eIto M, Sugito M, Kobayashi A, Nishizawa Y, Tsunoda Y, Saito N (2008) Relationship between multiple numbers of stapler firings during rectal division and anastomotic leakage after laparoscopic rectal resection. Int J Colorectal Dis 23(7):703-7. https://doi.org/10.1007/s00384-008-0470-8\u003c/li\u003e\n \u003cli\u003eZhang ZT (2019) Chinese Expert Consensus Statement on the Diagnostic, Prevention and Treation of the Anastomotic Leakage for Rectal Cancer (article in Chinese). Zhonghua Wei Chang Wai Ke Za Zhi 22(3):201-206. https://doi.org/10.3760/cma.j.issn.1671-0274.2019.03.001\u003c/li\u003e\n \u003cli\u003eBertani E, Chiappa A, Della Vigna P, Radice D, Papis D, Cossu L, et al. (2014) The Impact of pelvimetry on anastomotic leakage in a consecutive series of open, laparoscopic and robotic low anterior resections with total mesorectal excision for rectal cancer. Hepatogastroenterology 61(134):1574-81. https://doi.org/10.5754/hge13724\u003c/li\u003e\n \u003cli\u003eKilleen T, Banerjee S, Vijay V, Al-Dabbagh Z, Francis D, Warren S (2010) Magnetic resonance (MR) pelvimetry as a predictor of difficulty in laparoscopic operations for rectal cancer. Surgical endoscopy 24(12):2974-9. https://doi.org/10.1007/s00464-010-1075-1\u003c/li\u003e\n \u003cli\u003eZhou X, Su M, Hu K, Su Y, Ye Y, Huang C, et al. (2016) CT pelvimetry and clinicopathological parameters in evaluation of the technical difficulties in performing open rectal surgery for mid-low rectal cancer. Oncology letters 11(1):31-8. https://doi.org/10.3892/ol.2015.3827\u003c/li\u003e\n \u003cli\u003eBalyasnikova S, Brown G (2016) Optimal Imaging Strategies for Rectal Cancer Staging and Ongoing Management. Curr Treat Options Oncol 17(6):32. https://doi.org/10.1007/s11864-016-0403-7\u003c/li\u003e\n \u003cli\u003eJhaveri KS, Sadaf A (2009) Role of MRI for staging of rectal cancer. Expert Rev Anticancer Ther 9(4):469-81. https://doi.org/10.1586/era.09.13\u003c/li\u003e\n \u003cli\u003eRajjak S, Kureshi AK (2021) Multiple-Object Detection and Segmentation Based on Deep Learning in High-Resolution Video Using Mask-RCNN. International Journal of Pattern Recognition and Artificial Intelligence 35(13). https://doi.org/10.1142/S0218001421500385\u003c/li\u003e\n \u003cli\u003eTran D, Bourdev L, Fergus R, Torresani L, Paluri M (2015) Learning Spatiotemporal Features with 3D Convolutional Networks. IEEE International Conference on Computer Vision. pp, 4489-4497. https://doi.org/10.1109/ICCV.2015.510\u003c/li\u003e\n \u003cli\u003eLi W, Wang G, Fidon L, Ourselin S, Cardoso MJ, Vercauteren T (2017) On the Compactness, Efficiency, and Representation of 3D Convolutional Networks: Brain Parcellation as a Pretext Task. Springer, Cham 10265: 348\u0026ndash;360. https://doi.org/10.1007/978-3-319-59050-9_28\u003c/li\u003e\n \u003cli\u003eMeng J, Xue L, Chang Y, Zhang J, Yang K (2020) Automatic detection and segmentation of adenomatous colorectal polyps during colonoscopy using Mask R-CNN. Open Life Sciences 15(1):588-96. https://doi.org/10.1515/biol-2020-0055\u003c/li\u003e\n \u003cli\u003eQadir HA, Shin Y, Solhusvik J, Bergsland J, Balasingham I (2019) Polyp Detection and Segmentation using Mask R-CNN: Does a Deeper Feature Extractor CNN Always Perform Better? 2019 13th International Symposium on Medical Information and Communication Technology (ISMICT) pp, 1-6. https://doi.org/10.1109/ISMICT.2019.8743694\u003c/li\u003e\n \u003cli\u003eRussell BC, Torralba A, Murphy KP, Freeman WT (2008) LabelMe: A Database and Web-Based Tool for Image Annotation. International Journal of Computer Vision 77(1-3) : 157\u0026ndash;173. https://doi.org/10.1007/s11263-007-0090-8\u003c/li\u003e\n \u003cli\u003eLin TY, Maire M, Belongie S, Hays J, Zitnick CL (2014) Microsoft COCO: Common Objects in Context. Springer International Publishing. https://doi.org/10.1007/978-3-319-10602-1_48\u003c/li\u003e\n \u003cli\u003eHu J (2013) Proceedings of the 2013 IEEE International Conference on Computer Vision. IEEE International Conference on Computer Vision.\u003c/li\u003e\n \u003cli\u003eHuang MJ, Liang JL, Wang H, Kang L, Deng YH, Wang JP (2011) Laparoscopic-assisted versus open surgery for rectal cancer: a meta-analysis of randomized controlled trials on oncologic adequacy of resection and long-term oncologic outcomes. International Journal of Colorectal Disease 26(4):415. https://doi.org/10.1007/s00384-010-1091-6\u003c/li\u003e\n \u003cli\u003eKim C, Baek S, Hur H, Min B, Baik S, Kim N (2016) Anastomotic Leakage After Low Anterior Resection for Rectal Cancer Is Different Between Minimally Invasive Surgery and Open Surgery. Annals of surgery 263(1):130-7. https://doi.org/10.1097/sla.0000000000001157\u003c/li\u003e\n \u003cli\u003eKawada K, Hasegawa S, Hida K, Hirai K, Okoshi K, Nomura A, et al. (2014) Risk factors for anastomotic leakage after laparoscopic low anterior resection with DST anastomosis. Surgical Endoscopy 28(10):2988-95. https://doi.org/10.1007/s00464-014-3564-0\u003c/li\u003e\n \u003cli\u003eKuroyanagi H, Oya M, Ueno M, Fujimoto Y, Yamaguchi T, Muto T (2008) Standardized technique of laparoscopic intracorporeal rectal transection and anastomosis for low anterior resection. Surgical endoscopy 22(2):557-61. https://doi.org/10.1007/s00464-007-9626-9\u003c/li\u003e\n \u003cli\u003eNakagoe T, Ishikawa H, Sawai T, Tsuji T, Takeshita H, Nanashima A, et al. (2005) Oncological outcome of ultra-low anterior resection with total mesorectal excision for carcinoma of the lower third of the rectum: Comparison of intrapelvic double-stapled anastomosis and transanal coloanal anastomosis. Hepato-gastroenterology 52(66):1692-7. https://doi.org/10.1246/bcsj.40.2063\u003c/li\u003e\n \u003cli\u003eSpinelli A, Foppa C, Carvello M, Sacchi M, De Lucia F, Clerico G, et al. (2021) Transanal Transection and Single-Stapled Anastomosis (TTSS): A comparison of anastomotic leak rates with the double-stapled technique and with transanal total mesorectal excision (TaTME) for rectal cancer. European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology 47(12):3123-9. https://doi.org/10.1016/j.ejso.2021.08.002\u003c/li\u003e\n \u003cli\u003ePenna M, Hompes R, Arnold S, Wynn G, Austin R, Warusavitarne J, et al. (2019) Incidence and Risk Factors for Anastomotic Failure in 1594 Patients Treated by Transanal Total Mesorectal Excision: Results From the International TaTME Registry. Annals of surgery 269(4):700-11. https://doi.org/10.1097/sla.0000000000002653\u003c/li\u003e\n \u003cli\u003eFoo C, Hung H, Ho Y, Lam W, Law W (2020) Predicting the level of difficulty of the double-stapling technique in laparoscopic total mesorectal excision. Surgical endoscopy 34(8):3382-7. https://doi.org/10.1007/s00464-019-07112-2\u003c/li\u003e\n \u003cli\u003eLee S, Ahn B, Lee S (2017) The Relationship Between the Number of Intersections of Staple Lines and Anastomotic Leakage After the Use of a Double Stapling Technique in Laparoscopic Colorectal Surgery. Surgical laparoscopy, endoscopy \u0026amp; percutaneous techniques 27(4):273-81. https://doi.org/10.1097/sle.0000000000000422\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; Clinical characteristics of patients in deep-learning set\u003c/p\u003e\n\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"568\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eNumber of linear stapler firings\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026ge;3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026le;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003en = 58 (17.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003en = 270 (82.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003ep value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eAge [y]\u003c/p\u003e\n \u003cp\u003e＞70\u003c/p\u003e\n \u003cp\u003e\u0026le;70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e16 (27.6)\u003c/p\u003e\n \u003cp\u003e42 (72.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e54 (20.0)\u003c/p\u003e\n \u003cp\u003e216 (80.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.218\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eGender, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.084\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e46 (79.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e181 (67.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e12 (20.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e89 (33.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eBMI [Kg/m2]\u003c/p\u003e\n \u003cp\u003e＞25\u003c/p\u003e\n \u003cp\u003e\u0026le;25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e17 (29.3)\u003c/p\u003e\n \u003cp\u003e41 (70.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e72 (26.7)\u003c/p\u003e\n \u003cp\u003e198 (73.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.745\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eDiabetes mellitus, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.873\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e9 (15.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e32 (11.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e49 (84.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e238 (88.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003enCRT, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.709\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e14 (24.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e71 (26.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e44 (75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e199 (73.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eOperation time [min]\u003c/p\u003e\n \u003cp\u003e\u0026gt;150\u003c/p\u003e\n \u003cp\u003e\u0026le;150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e32 (55.2)\u003c/p\u003e\n \u003cp\u003e26 (44.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e124 (45.9)\u003c/p\u003e\n \u003cp\u003e146 (54.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.246\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eAnastomotic leakage, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e14 (24.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e34 (12.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e44 (75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e236 (87.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBiochemical data\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eHemoglobin [g/L]\u003c/p\u003e\n \u003cp\u003e\u0026lt;120\u003c/p\u003e\n \u003cp\u003e\u0026ge;120\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e12 (20.7)\u003c/p\u003e\n \u003cp\u003e46 (79.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e58 (21.5)\u003c/p\u003e\n \u003cp\u003e212 (78.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.893\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eAlbumin [g/L]\u003c/p\u003e\n \u003cp\u003e\u0026lt;35\u003c/p\u003e\n \u003cp\u003e\u0026ge;35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e8 (13.8)\u003c/p\u003e\n \u003cp\u003e50 (86.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e22 (8.1)\u003c/p\u003e\n \u003cp\u003e248 (91.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.207\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eCEA [ng/mL]\u003c/p\u003e\n \u003cp\u003e\u0026gt;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e26 (44.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e71 (26.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003e\u0026le;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e32 (55.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e199 (73.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003e\u003cstrong\u003eTumor characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eDistance from anus [cm]\u003c/p\u003e\n \u003cp\u003e\u0026lt;5\u003c/p\u003e\n \u003cp\u003e\u0026ge;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e4 (6.9)\u003c/p\u003e\n \u003cp\u003e54 (93.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e39 (14.4)\u003c/p\u003e\n \u003cp\u003e231 (85.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.138\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eTumor size [cm]\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026ge;5\u003c/p\u003e\n \u003cp\u003e\u0026lt;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e14 (24.1)\u003c/p\u003e\n \u003cp\u003e44 (75.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e26 (9.6)\u003c/p\u003e\n \u003cp\u003e244 (90.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eT stage, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.865\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eT\u0026le;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e14 (24.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e68 (25.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eT3-4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e44 (75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e229 (84.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eN stage, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.663\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eN0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e30(51.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e152 (56.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eN+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e28(48.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e118 (43.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eCRM (evaluated by MRI), n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003ePositive\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e20 (34.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e52 (19.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.24647887323944%\"\u003e\n \u003cp\u003eNegative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.887323943661972%\"\u003e\n \u003cp\u003e38 (65.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.654929577464788%\"\u003e\n \u003cp\u003e218 (80.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.211267605633802%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviation: MRI: magnetic resonance imaging; CRM: circumferential resection margin; N: node; T: tumor; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index.\u003c/p\u003e\n\n\n\u003cp\u003eTable 2 Risk factors of \u0026ge;3 linear stapler firings in deep-learning set\u003c/p\u003e\n\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"558\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.18850987432675%\"\u003e\n \u003cp\u003eFactors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" width=\"32.85457809694793%\"\u003e\n \u003cp\u003eUnivariate analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" width=\"31.956912028725313%\"\u003e\n \u003cp\u003eMultivariate analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\u0026nbsp;\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003eOR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003ep value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eOR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003ep value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eAge [y] (\u0026ge;70/\u0026lt;70)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e1.45 (0.82, 2.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.202\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eGender (Male/Female)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e1.89 (0.95, 3.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.069\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003e1.82 (0.90, 3.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.098\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eBMI [Kg/m2] (\u0026ge;25/\u0026lt;25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e1.14 (0.61, 2.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.681\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eDiabetes mellitus (Y/N)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e1.37 (0.61, 3.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.445\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003enCRT (Y/N)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e0.89 (0.46, 1.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.734\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eAlbumin [g/L] (\u0026lt;35/\u0026ge;35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e1.80 (0.76, 4.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.181\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eCEA [ng/mL] (\u0026gt;5/\u0026le;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e2.39 (1.33, 4.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003e2.20 (1.20, 4.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.011\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eDistance from anus [cm] (\u0026lt;5/\u0026ge;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e2.28 (0.78, 6.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003eN.A.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eTumor size [cm] (\u0026ge;5/\u0026lt;5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e2.99 (1.45, 6.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003e2.54 (1.15, 5.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"35.12544802867384%\"\u003e\n \u003cp\u003eCRM (evaluated by MRI) (+/-)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"22.40143369175627%\"\u003e\n \u003cp\u003e2.21 (1.19, 4.10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.012\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"21.684587813620073%\"\u003e\n \u003cp\u003e1.66 (0.84, 3.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.394265232974911%\"\u003e\n \u003cp\u003e0.143\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviation:\u0026nbsp;MRI: magnetic resonance imaging; CRM: circumferential resection margin; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index; N.A.: not applicable; OR: odds ratio; CI: confidence interval.\u003c/p\u003e\n\n\n\u003cp\u003eTable 3 \u0026nbsp;Comparison of clinical characteristics between deep-learning set and validation set\u003c/p\u003e\n\u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"558\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003eDeep-learning set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003eValidation set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003ep value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003en = 328\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003en=128\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eAge [y]\u003c/p\u003e\n \u003cp\u003e＞70\u003c/p\u003e\n \u003cp\u003e\u0026le;70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e70 (21.3)\u003c/p\u003e\n \u003cp\u003e258 (78.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e36 (28.1)\u003c/p\u003e\n \u003cp\u003e92 (71.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.139\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eGender, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.561\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e227 (69.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e83 (64.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e101 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e45 (35.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eBMI [Kg/m2]\u003c/p\u003e\n \u003cp\u003e＞25\u003c/p\u003e\n \u003cp\u003e\u0026le;25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e89 (27.1)\u003c/p\u003e\n \u003cp\u003e239 (72.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e42 (32.8)\u003c/p\u003e\n \u003cp\u003e86 (67.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.250\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eDiabetes mellitus, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.342\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e41 (12.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e23 (18.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e287 (87.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e105 (82.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003enCRT, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.201\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e85 (25.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e41 (32.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e243 (74.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e87 (68.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eNumber of linear stapler cartridges, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.204\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026ge;3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e58 (17.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e16 (12.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026le;2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e270 (82.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e112 (87.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eAnastomotic leakage, n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.090\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e48 (14.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e11 (8.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e280 (85.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e117 (91.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBiochemical data\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eAlbumin [g/L]\u003c/p\u003e\n \u003cp\u003e\u0026lt;35\u003c/p\u003e\n \u003cp\u003e\u0026ge;35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e97 (29.6)\u003c/p\u003e\n \u003cp\u003e231 (70.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e32 (25.0)\u003c/p\u003e\n \u003cp\u003e96 (75.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.356\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eCEA [ng/mL], n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e(Missing=5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.301\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026gt;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e97 (29.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e32 (25.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026le;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e226 (68.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e96 (75.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u003cstrong\u003eTumor characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eDistance from anus [cm]\u003c/p\u003e\n \u003cp\u003e\u0026lt;5\u003c/p\u003e\n \u003cp\u003e\u0026ge;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e43 (13.1)\u003c/p\u003e\n \u003cp\u003e285 (86.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e14 (10.9)\u003c/p\u003e\n \u003cp\u003e114 (89.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.637\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eTumor size [cm], n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.130\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026ge;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e40 (12.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e9 (7.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003e\u0026lt;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e288 (87.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e119 (93.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eCRM (evaluated by MRI), n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e0.092\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003ePositive\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e72 (22.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e19 (14.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"45.78096947935368%\"\u003e\n \u003cp\u003eNegative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"18.671454219030522%\"\u003e\n \u003cp\u003e256 (78.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"23.6983842010772%\"\u003e\n \u003cp\u003e109 (69.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.8491921005386%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviation: MRI: magnetic resonance imaging; CRM: circumferential resection margin; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Low anterior resection (LAR), Double stapling technique (DST), Rectal cancer, Magnetic resonance imaging (MRI), Deep-learning, Predictive model","lastPublishedDoi":"10.21203/rs.3.rs-2681419/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2681419/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMultiple linear stapler firings is a risk factor for anastomotic leakage (AL) in laparoscopic low anterior resection (LAR) using double stapling technique (DST) anastomosis. In this study, our objective was to establish the risk factors for ≥3 linear stapler firings, and to create and validate a predictive model for ≥3 linear stapler firings in laparoscopic LAR using DST anastomosis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe retrospectively enrolled 328 mid–low rectal cancer patients undergoing laparoscopic LAR using DST anastomosis. With a split ratio of 4:1, patients were randomly divided into 2 sets: the training set (n = 260) and the testing set (n = 68). A clinical predictive model of ≥3 linear stapler firings was constructed by binary logistic regression. Based on three-dimensional convolutional networks, we built an image model using only magnetic resonance (MR) images segmented by Mask region-based convolutional neural network, and an integrated model based on both MR images and clinical variables. Area under the curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV), and Youden index were calculated for each model. And the three models were externally validated by an independent cohort of 128 patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere were 17.7% (58/328) patients received ≥3 linear stapler firings. Tumor size ≥5 cm (odds ratio (OR)=2.54, 95% confidence interval (CI)=1.15–5.60, p=0.021) and preoperative carcinoma embryonic antigen (CEA) level \u0026gt;5 ng/mL [OR=2.20, 95% CI=1.20–4.04, p=0.011] were independent risk factors associated with ≥3 linear stapler firings. The integrated model (AUC=0.88, accuracy=94.1%) performed better on predicting ≥3 linear stapler firings than the clinical model (AUC =0.72, accuracy=86.7%) and the image model (AUC=0.81, accuracy=91.2%). Similarly, in the validation set, the integrated model (AUC=0.84, accuracy=93.8%) performed better than the clinical model (AUC =0.65, accuracy=65.6%) and the image model (AUC=0.75, accuracy=92.1%).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur deep-learning model based on pelvic MR can help predict the high-risk population with ≥3 linear stapler firings in laparoscopic LAR using DST anastomosis. This model might assist in determining preoperatively the anastomotic technique for mid–low rectal cancer patients.\u003c/p\u003e","manuscriptTitle":"Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-03-29 14:22:08","doi":"10.21203/rs.3.rs-2681419/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2023-09-23T05:20:42+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2023-09-21T11:00:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"4e0520b8-0f44-4933-9638-897bf768a3f3_SNPRID","date":"2023-09-11T22:03:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2023-04-10T13:51:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"1969df4c-f72e-426a-9982-afacadb32709","date":"2023-04-09T22:57:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"1eacaf38-59b0-47ec-9a4c-80b0242a5ed8","date":"2023-04-08T10:46:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2023-04-08T10:35:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-03-27T14:22:18+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2023-03-27T11:24:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2023-03-27T11:13:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2023-03-11T14:40:56+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b6aced52-ce81-4bd1-ab04-9d1d424abc6a","owner":[],"postedDate":"March 29th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":20232788,"name":"Biological sciences/Cancer"},{"id":20232789,"name":"Health sciences/Gastroenterology"},{"id":20232790,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2023-11-06T15:05:47+00:00","versionOfRecord":{"articleIdentity":"rs-2681419","link":"https://doi.org/10.1038/s41598-023-46225-6","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2023-11-02 15:00:43","publishedOnDateReadable":"November 2nd, 2023"},"versionCreatedAt":"2023-03-29 14:22:08","video":"","vorDoi":"10.1038/s41598-023-46225-6","vorDoiUrl":"https://doi.org/10.1038/s41598-023-46225-6","workflowStages":[]},"version":"v1","identity":"rs-2681419","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2681419","identity":"rs-2681419","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}