Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients

PurposeThis study aimed to develop a machine learning (ML) model for real-time prediction of duodenal stump leakage (DSL) following gastrectomy in patients with gastric cancer (GC) using a comprehensive set of clinical variables to improve postoperative outcomes and monitoring efficiency.MethodsA re...

Full description

Saved in:
Bibliographic Details
Main Authors: Jae Hun Chung, Yushin Kim, Dongjun Lee, Dongwon Lim, Sun-Hwi Hwang, Si-Hak Lee, Woohwan Jung
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Surgery
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fsurg.2025.1550990/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850279947117002752
author Jae Hun Chung
Jae Hun Chung
Jae Hun Chung
Yushin Kim
Dongjun Lee
Dongwon Lim
Dongwon Lim
Dongwon Lim
Sun-Hwi Hwang
Sun-Hwi Hwang
Sun-Hwi Hwang
Si-Hak Lee
Si-Hak Lee
Si-Hak Lee
Woohwan Jung
author_facet Jae Hun Chung
Jae Hun Chung
Jae Hun Chung
Yushin Kim
Dongjun Lee
Dongwon Lim
Dongwon Lim
Dongwon Lim
Sun-Hwi Hwang
Sun-Hwi Hwang
Sun-Hwi Hwang
Si-Hak Lee
Si-Hak Lee
Si-Hak Lee
Woohwan Jung
author_sort Jae Hun Chung
collection DOAJ
description PurposeThis study aimed to develop a machine learning (ML) model for real-time prediction of duodenal stump leakage (DSL) following gastrectomy in patients with gastric cancer (GC) using a comprehensive set of clinical variables to improve postoperative outcomes and monitoring efficiency.MethodsA retrospective analysis was conducted on 1,107 patients with GC who underwent gastrectomy at Pusan National University Yangsan Hospital between 2019 and 2022. One hundred eighty-nine features were extracted from each patient record, including demographic data, preoperative comorbidities, and blood test outcomes from the subsequent seven postoperative days (POD). Six ML algorithms were evaluated: Logistic Regression (LR), K-nearest neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), Extreme Gradient Boosting (XGB), and Neural Network (NN). The models predicted DSL occurrence preoperatively and on POD 1, 2, 3, 5, and 7. Performance was assessed using the Area Under the Receiver Operating Characteristic Curve (AUROC) and Recall@K.ResultsAmong the 1,107 patients, 29 developed DSL. XGB demonstrated the highest AUROC score (0.880), followed by RF (0.858), LR (0.823), SVM (0.819), NN (0.753), and KNN (0.726). The RF achieved the best Recall@K score of 0.643. Including additional POD features improved the predictive performance, with the AUROC value increasing to 0.879 on POD 7. The confidence scores of the model indicated that the DSL predictions became more reliable over time.ConclusionThe study concluded that ML models, notably the XGB algorithm, can effectively predict DSL in real-time using comprehensive clinical data, enhancing the clinical decision-making process for GC patients.
format Article
id doaj-art-2f1f24c1c9a84ea19cfafcc20d6a3821
institution OA Journals
issn 2296-875X
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Surgery
spelling doaj-art-2f1f24c1c9a84ea19cfafcc20d6a38212025-08-20T01:48:57ZengFrontiers Media S.A.Frontiers in Surgery2296-875X2025-05-011210.3389/fsurg.2025.15509901550990Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patientsJae Hun Chung0Jae Hun Chung1Jae Hun Chung2Yushin Kim3Dongjun Lee4Dongwon Lim5Dongwon Lim6Dongwon Lim7Sun-Hwi Hwang8Sun-Hwi Hwang9Sun-Hwi Hwang10Si-Hak Lee11Si-Hak Lee12Si-Hak Lee13Woohwan Jung14Division of Gastrointestinal Surgery, Department of Surgery, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaResearch Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaDepartment of Surgery, School of Medicine, Pusan National University, Yangsan, Republic of KoreaDepartment of Applied Artificial Intelligence (Major in Bio Artificial Intelligence), Hanyang University, Ansan, Republic of KoreaDepartment of Applied Artificial Intelligence (Major in Bio Artificial Intelligence), Hanyang University, Ansan, Republic of KoreaDivision of Gastrointestinal Surgery, Department of Surgery, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaResearch Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaDepartment of Surgery, School of Medicine, Pusan National University, Yangsan, Republic of KoreaDivision of Gastrointestinal Surgery, Department of Surgery, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaResearch Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaDepartment of Surgery, School of Medicine, Pusan National University, Yangsan, Republic of KoreaDivision of Gastrointestinal Surgery, Department of Surgery, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaResearch Institute for Convergence of Biomedical Science and Technology, Pusan National University Yangsan Hospital, Yangsan, Republic of KoreaDepartment of Surgery, School of Medicine, Pusan National University, Yangsan, Republic of KoreaDepartment of Artificial Intelligence, Hanyang University, Ansan, Republic of KoreaPurposeThis study aimed to develop a machine learning (ML) model for real-time prediction of duodenal stump leakage (DSL) following gastrectomy in patients with gastric cancer (GC) using a comprehensive set of clinical variables to improve postoperative outcomes and monitoring efficiency.MethodsA retrospective analysis was conducted on 1,107 patients with GC who underwent gastrectomy at Pusan National University Yangsan Hospital between 2019 and 2022. One hundred eighty-nine features were extracted from each patient record, including demographic data, preoperative comorbidities, and blood test outcomes from the subsequent seven postoperative days (POD). Six ML algorithms were evaluated: Logistic Regression (LR), K-nearest neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), Extreme Gradient Boosting (XGB), and Neural Network (NN). The models predicted DSL occurrence preoperatively and on POD 1, 2, 3, 5, and 7. Performance was assessed using the Area Under the Receiver Operating Characteristic Curve (AUROC) and Recall@K.ResultsAmong the 1,107 patients, 29 developed DSL. XGB demonstrated the highest AUROC score (0.880), followed by RF (0.858), LR (0.823), SVM (0.819), NN (0.753), and KNN (0.726). The RF achieved the best Recall@K score of 0.643. Including additional POD features improved the predictive performance, with the AUROC value increasing to 0.879 on POD 7. The confidence scores of the model indicated that the DSL predictions became more reliable over time.ConclusionThe study concluded that ML models, notably the XGB algorithm, can effectively predict DSL in real-time using comprehensive clinical data, enhancing the clinical decision-making process for GC patients.https://www.frontiersin.org/articles/10.3389/fsurg.2025.1550990/fullduodenal stump leakagegastrectomymachine learninggastric cancerpredictive modeling
spellingShingle Jae Hun Chung
Jae Hun Chung
Jae Hun Chung
Yushin Kim
Dongjun Lee
Dongwon Lim
Dongwon Lim
Dongwon Lim
Sun-Hwi Hwang
Sun-Hwi Hwang
Sun-Hwi Hwang
Si-Hak Lee
Si-Hak Lee
Si-Hak Lee
Woohwan Jung
Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
Frontiers in Surgery
duodenal stump leakage
gastrectomy
machine learning
gastric cancer
predictive modeling
title Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
title_full Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
title_fullStr Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
title_full_unstemmed Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
title_short Machine learning-based real-time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
title_sort machine learning based real time prediction of duodenal stump leakage from gastrectomy in gastric cancer patients
topic duodenal stump leakage
gastrectomy
machine learning
gastric cancer
predictive modeling
url https://www.frontiersin.org/articles/10.3389/fsurg.2025.1550990/full
work_keys_str_mv AT jaehunchung machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT jaehunchung machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT jaehunchung machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT yushinkim machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT dongjunlee machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT dongwonlim machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT dongwonlim machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT dongwonlim machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sunhwihwang machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sunhwihwang machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sunhwihwang machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sihaklee machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sihaklee machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT sihaklee machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients
AT woohwanjung machinelearningbasedrealtimepredictionofduodenalstumpleakagefromgastrectomyingastriccancerpatients