Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations
Introduction: Lung cancer is a prevalent malignancy globally, with approximately 20% of patients developing cardiopulmonary complications after lobectomy. In order to prevent complications, an accurate and personalized method based on machine learning (ML) is required. Methods: During the period of...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-07-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023049800 |
| _version_ | 1797771490916040704 |
|---|---|
| author | Yihai Zhai Xue Lin Qiaolin Wei Yuanjin Pu Yonghui Pang |
| author_facet | Yihai Zhai Xue Lin Qiaolin Wei Yuanjin Pu Yonghui Pang |
| author_sort | Yihai Zhai |
| collection | DOAJ |
| description | Introduction: Lung cancer is a prevalent malignancy globally, with approximately 20% of patients developing cardiopulmonary complications after lobectomy. In order to prevent complications, an accurate and personalized method based on machine learning (ML) is required. Methods: During the period of 2017–2021, a retrospective analysis was conducted on the medical records of patients who had undergone lobectomy for non-small cell lung cancer (NSCLC). We performed logical regression, decision tree (DT), random forest (RF), gradient boost DT, and eXtreme gradient boosting analyses to establish an ML model. The ten-fold cross-validation was used to evaluate the performance of multiple ML models based on various evaluation metrics, including accuracy, precision, recall, F1 score, and area under the receiver operating (AUC). Additionally, we also calculated the Kappa value of these model. Each model used grid search to optimize hyper-parameters and then used the interpretability method to provide explanations for the model's Decisions. Results: The study included 718 eligible patients, among whom the incidence of postoperative cardiopulmonary complications was 20.89%. The RF model showed the best comprehensive performance among all models, and its ten-fold cross-validation accuracy, precision, recall, F1 score, and AUC were (OR and 95% confidence interval [CI]) 0.786 (0.738–0.834), 0.803 (0.735–0.872), 0.738 (0.678–0.797), 0.766 (0.714–0.818), 0.856 (0.815–0.898), respectively. The kappa value of the RF model was 0.696 (0.617–0.768). The SHAP method showed that gender, age, and intraoperative blood loss were closely associated with postoperative cardiopulmonary complications. Conclusion: The application of ML methods for predicting postoperative cardiopulmonary complications based on clinical data of patients with NSCLC showed a good performance. The results indicate that ML combined with the SHAP individualized interpretation method has practical clinical value. |
| first_indexed | 2024-03-12T21:38:21Z |
| format | Article |
| id | doaj.art-4e53881bef494da3b9cb0a48dd57b34c |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-03-12T21:38:21Z |
| publishDate | 2023-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-4e53881bef494da3b9cb0a48dd57b34c2023-07-27T05:57:23ZengElsevierHeliyon2405-84402023-07-0197e17772Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanationsYihai Zhai0Xue Lin1Qiaolin Wei2Yuanjin Pu3Yonghui Pang4Guangxi Medical University Cancer Hospital, Department of Thoracic Surgery, Nanning, 530021, ChinaThe Second Affiliated Hospital of Guangxi Medical University, Department of Oncology, Nanning, 530000, ChinaGuangxi Medical University Cancer Hospital, Department of Interventional Therapy, Nanning, 530021, ChinaGuangxi Medical University Cancer Hospital, Department of Thoracic Surgery, Nanning, 530021, ChinaGuangxi Medical University Cancer Hospital, Department of Thoracic Surgery, Nanning, 530021, China; Corresponding author.Introduction: Lung cancer is a prevalent malignancy globally, with approximately 20% of patients developing cardiopulmonary complications after lobectomy. In order to prevent complications, an accurate and personalized method based on machine learning (ML) is required. Methods: During the period of 2017–2021, a retrospective analysis was conducted on the medical records of patients who had undergone lobectomy for non-small cell lung cancer (NSCLC). We performed logical regression, decision tree (DT), random forest (RF), gradient boost DT, and eXtreme gradient boosting analyses to establish an ML model. The ten-fold cross-validation was used to evaluate the performance of multiple ML models based on various evaluation metrics, including accuracy, precision, recall, F1 score, and area under the receiver operating (AUC). Additionally, we also calculated the Kappa value of these model. Each model used grid search to optimize hyper-parameters and then used the interpretability method to provide explanations for the model's Decisions. Results: The study included 718 eligible patients, among whom the incidence of postoperative cardiopulmonary complications was 20.89%. The RF model showed the best comprehensive performance among all models, and its ten-fold cross-validation accuracy, precision, recall, F1 score, and AUC were (OR and 95% confidence interval [CI]) 0.786 (0.738–0.834), 0.803 (0.735–0.872), 0.738 (0.678–0.797), 0.766 (0.714–0.818), 0.856 (0.815–0.898), respectively. The kappa value of the RF model was 0.696 (0.617–0.768). The SHAP method showed that gender, age, and intraoperative blood loss were closely associated with postoperative cardiopulmonary complications. Conclusion: The application of ML methods for predicting postoperative cardiopulmonary complications based on clinical data of patients with NSCLC showed a good performance. The results indicate that ML combined with the SHAP individualized interpretation method has practical clinical value.http://www.sciencedirect.com/science/article/pii/S2405844023049800Non–small cell lung cancerComplicationsRandom forestInterpretable model |
| spellingShingle | Yihai Zhai Xue Lin Qiaolin Wei Yuanjin Pu Yonghui Pang Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations Heliyon Non–small cell lung cancer Complications Random forest Interpretable model |
| title | Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations |
| title_full | Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations |
| title_fullStr | Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations |
| title_full_unstemmed | Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations |
| title_short | Interpretable prediction of cardiopulmonary complications after non-small cell lung cancer surgery based on machine learning and SHapley additive exPlanations |
| title_sort | interpretable prediction of cardiopulmonary complications after non small cell lung cancer surgery based on machine learning and shapley additive explanations |
| topic | Non–small cell lung cancer Complications Random forest Interpretable model |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023049800 |
| work_keys_str_mv | AT yihaizhai interpretablepredictionofcardiopulmonarycomplicationsafternonsmallcelllungcancersurgerybasedonmachinelearningandshapleyadditiveexplanations AT xuelin interpretablepredictionofcardiopulmonarycomplicationsafternonsmallcelllungcancersurgerybasedonmachinelearningandshapleyadditiveexplanations AT qiaolinwei interpretablepredictionofcardiopulmonarycomplicationsafternonsmallcelllungcancersurgerybasedonmachinelearningandshapleyadditiveexplanations AT yuanjinpu interpretablepredictionofcardiopulmonarycomplicationsafternonsmallcelllungcancersurgerybasedonmachinelearningandshapleyadditiveexplanations AT yonghuipang interpretablepredictionofcardiopulmonarycomplicationsafternonsmallcelllungcancersurgerybasedonmachinelearningandshapleyadditiveexplanations |