Development and validation of artificial intelligence models for preoperative prediction of inferior mesenteric artery lymph nodes metastasis in left colon and rectal cancer |
| |
| Institution: | 1. Department of Colorectal Surgery, Union Hospital, Fujian Medical University, People''s Republic of China;2. Department of Pathology, Union Hospital, Fujian Medical University, People''s Republic of China;1. Gynaecologic Oncology Unit, Vall d''Hebron University Hospital, Barcelona, Spain;2. Health Services Research Group, Vall d’Hebron Institut de Recerca (VHIR), Barcelona, Spain;3. Information Systems, Vall d’Hebron University Hospital, Barcelona, Spain;4. Biomedical Research Group in Gynaecology, Vall d''Hebron Research Institute (VHIR), Universitat Autonoma de Barcelona, CIBERONC, Barcelona, Spain;1. Digestive and General Surgery, Yalgado Ouedraogo Teaching Hospital Ouagadougou, 03 BP 7021, Ouagadougou, Joseph Ki-Zerbo University, Burkina Faso;2. Public Health Department, Yalgado Ouedraogo Teaching Hospital Ouagadougou, 03 BP 7021, Ouagadougou, Joseph Ki-Zerbo University, Burkina Faso;3. Institut Joliot Curie de Dakar (Sénégal), Cheikh Anta Diop University of Dakar, 10700, Dakar, Senegal;1. Key Laboratory of Integrated Oncology and Intelligent Medicine of Zhejiang Province, Department of Hepatobiliary and Pancreatic Surgery, Affiliated Hangzhou First People''s Hospital, Zhejiang University School of Medicine, Hangzhou, 310006, China;2. Westlake Laboratory of Life Sciences and Biomedicine, Hangzhou, 310024, China;3. NHC Key Laboratory of Combined Multi-organ Transplantation, Hangzhou, 310003, China;4. Institute of Organ Transplantation, Zhejiang University, Hangzhou, 310003, China;1. Department of Hematology and Oncology, Asklepios Hospital Barmbek, Hamburg, Germany;2. Asklepios Campus Hamburg, Semmelweis University, Budapest, Hungary;3. Department of General and Abdominal Surgery, Asklepios Hospital Barmbek, Hamburg, Germany;4. Surgical Clinic Unit, Department of Surgery and Medical Surgical Specialties, University of Catania, Italy;1. Department of Research and Development, Netherlands Comprehensive Cancer Organisation (IKNL), Utrecht, the Netherlands;2. Department of Surgery, Erasmus Medical Center, Rotterdam, Netherlands;3. Department of Surgery, Catharina Hospital, Eindhoven, the Netherlands;4. Department of Public Health, Erasmus MC, Rotterdam, the Netherlands |
| |
| Abstract: | BackgroundDissection of lymph nodes at the roots of the inferior mesenteric artery (IMAN) should be offered only to selected patients at a major risk of developing IMAN involvement. The aim of this study is to present the first artificial intelligence (AI) models to predict IMAN metastasis risk in the left colon and rectal cancer patients.MethodsA total of 2891 patients with descending colon including splenic flexure, sigmoid colon and rectal cancer undergoing major primary surgery and IMAN dissection were included as a study cohort, which was then split into a training set (67%) and a testing set (33%). Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) regression model. Seven AI algorithms, namely Support Vector Machine (SVM), Logistic Regression (LR), Extreme Gradient Boosting (XGB), Light Gradient Boosting (LGB), Decision Tree Classifier (DTC), Random Forest (RF) classifier, and Multilayer Perceptron (MLP), as well as traditional multivariate LR model were employed to construct predictive models. The optimal hyperparameters were determined with 5 fold cross-validation. The predictive performance of models and the expert surgeon was assessed and compared in the testing set independently.ResultsThe IMAN involvement incidence was 4.6%. The optimal set of features selected by LASSO included 10 characteristics: neoadjuvant treatment, age, synchronous liver metastasis, synchronous lung metastasis, signet ring adenocarcinoma, neural invasion, lymphovascular invasion, CA199, endoscopic obstruction, T stage evaluated by MRI. The most accurate model derived from MLP showed excellent prediction power with area under the receiver operating characteristic curve (AUROC) of 0.873 and produced 81.0% recognition sensitivity and 82.5% specificity in the testing set independently. In contrast, the judgment of IMAN metastasis by expert surgeon yield rather imprecise and unreliable results with a significantly lower AUROC of 0.509. Additionally, the proposed MLP had the highest net benefits and the largest reduction of unnecessary IMAN dissection without the cost of additional involved IMAN missed.ConclusionMLP model was able to maintain its prediction accuracy in the testing set better than other models and expert surgeons. Our MLP model could be used to help identify IMA nodal metastasis and to select candidates for individual IMAN dissection. |
| |
| Keywords: | Inferior mesenteric artery lymph node Colorectal cancer Machine learning Multilayer perceptron |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
