Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma
Abstract Aim Lymph node metastasis is an adverse prognostic factor in pancreatic ductal adenocarcinoma. However, it remains a challenge to predict lymph node metastasis using preoperative imaging alone. We used machine learning (combining preoperative imaging findings, tumor markers, and clinical in...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-01-01
|
| Series: | Annals of Gastroenterological Surgery |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/ags3.12836 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846097134656421888 |
|---|---|
| author | Susumu Daibo Yuki Homma Hiroki Ohya Hironori Fukuoka Kentaro Miyake Mayumi Ozawa Takafumi Kumamoto Ryusei Matsuyama Yusuke Saigusa Itaru Endo |
| author_facet | Susumu Daibo Yuki Homma Hiroki Ohya Hironori Fukuoka Kentaro Miyake Mayumi Ozawa Takafumi Kumamoto Ryusei Matsuyama Yusuke Saigusa Itaru Endo |
| author_sort | Susumu Daibo |
| collection | DOAJ |
| description | Abstract Aim Lymph node metastasis is an adverse prognostic factor in pancreatic ductal adenocarcinoma. However, it remains a challenge to predict lymph node metastasis using preoperative imaging alone. We used machine learning (combining preoperative imaging findings, tumor markers, and clinical information) to create a novel prediction model for lymph node metastasis in resectable pancreatic ductal adenocarcinoma. Methods The data of patients with resectable pancreatic ductal adenocarcinoma who underwent surgery between September 1991 and October 2022 were retrospectively examined. Machine‐learning software (Statistical Package for the Social Sciences Modeler) was used to create a prediction model, and parameter tuning was performed to improve the model's accuracy. We also analyzed the contribution of each feature to prediction using individual conditional expectation and partial dependence plots. Results Of the 331 cases included in the study, 241 comprised the training cohort and 90 comprised the test cohort. After parameter tuning, the areas under the receiver operating characteristic curves for the training and test cohorts were 0.780 and 0.795, respectively. Individual conditional expectation and partial dependence plots showed that larger tumor size and carbohydrate antigen 19–9 and Duke pancreatic monoclonal antigen type 2 levels were associated with positive lymph node metastasis prediction in this model; neoadjuvant treatment was associated with negative lymph node metastasis prediction. Conclusion Machine learning may contribute to the creation of an effective predictive model of lymph node metastasis in pancreatic ductal adenocarcinoma. Prediction models using machine learning may contribute to the development of new treatment strategies in resectable pancreatic ductal adenocarcinoma. |
| format | Article |
| id | doaj-art-f3cf000936ab43a3adcc4e5b8883dab1 |
| institution | Kabale University |
| issn | 2475-0328 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Annals of Gastroenterological Surgery |
| spelling | doaj-art-f3cf000936ab43a3adcc4e5b8883dab12025-01-02T04:49:00ZengWileyAnnals of Gastroenterological Surgery2475-03282025-01-019116116810.1002/ags3.12836Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinomaSusumu Daibo0Yuki Homma1Hiroki Ohya2Hironori Fukuoka3Kentaro Miyake4Mayumi Ozawa5Takafumi Kumamoto6Ryusei Matsuyama7Yusuke Saigusa8Itaru Endo9Department of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Surgery, Gastroenterological Center Yokohama City University Medical Center Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanDepartment of Biostatistics Yokohama City University Yokohama Kanagawa JapanDepartment of Gastroenterological Surgery Yokohama City University Yokohama Kanagawa JapanAbstract Aim Lymph node metastasis is an adverse prognostic factor in pancreatic ductal adenocarcinoma. However, it remains a challenge to predict lymph node metastasis using preoperative imaging alone. We used machine learning (combining preoperative imaging findings, tumor markers, and clinical information) to create a novel prediction model for lymph node metastasis in resectable pancreatic ductal adenocarcinoma. Methods The data of patients with resectable pancreatic ductal adenocarcinoma who underwent surgery between September 1991 and October 2022 were retrospectively examined. Machine‐learning software (Statistical Package for the Social Sciences Modeler) was used to create a prediction model, and parameter tuning was performed to improve the model's accuracy. We also analyzed the contribution of each feature to prediction using individual conditional expectation and partial dependence plots. Results Of the 331 cases included in the study, 241 comprised the training cohort and 90 comprised the test cohort. After parameter tuning, the areas under the receiver operating characteristic curves for the training and test cohorts were 0.780 and 0.795, respectively. Individual conditional expectation and partial dependence plots showed that larger tumor size and carbohydrate antigen 19–9 and Duke pancreatic monoclonal antigen type 2 levels were associated with positive lymph node metastasis prediction in this model; neoadjuvant treatment was associated with negative lymph node metastasis prediction. Conclusion Machine learning may contribute to the creation of an effective predictive model of lymph node metastasis in pancreatic ductal adenocarcinoma. Prediction models using machine learning may contribute to the development of new treatment strategies in resectable pancreatic ductal adenocarcinoma.https://doi.org/10.1002/ags3.12836adenocarcinomaearly diagnosislymphatic metastasismachine learningpancreatic neoplasms |
| spellingShingle | Susumu Daibo Yuki Homma Hiroki Ohya Hironori Fukuoka Kentaro Miyake Mayumi Ozawa Takafumi Kumamoto Ryusei Matsuyama Yusuke Saigusa Itaru Endo Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma Annals of Gastroenterological Surgery adenocarcinoma early diagnosis lymphatic metastasis machine learning pancreatic neoplasms |
| title | Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| title_full | Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| title_fullStr | Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| title_full_unstemmed | Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| title_short | Novel machine‐learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| title_sort | novel machine learning model for predicting lymph node metastasis in resectable pancreatic ductal adenocarcinoma |
| topic | adenocarcinoma early diagnosis lymphatic metastasis machine learning pancreatic neoplasms |
| url | https://doi.org/10.1002/ags3.12836 |
| work_keys_str_mv | AT susumudaibo novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT yukihomma novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT hirokiohya novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT hironorifukuoka novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT kentaromiyake novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT mayumiozawa novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT takafumikumamoto novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT ryuseimatsuyama novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT yusukesaigusa novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma AT itaruendo novelmachinelearningmodelforpredictinglymphnodemetastasisinresectablepancreaticductaladenocarcinoma |