DOI 10.1007/s12672-026-05389-z
Machine learning for predicting liver metastasis in colorectal cancer.
Аннотация
AIM: To evaluate the performance of machine learning models in predicting liver metastasis in colorectal cancer (CRC) patients using the SEER database and external validation from Ningbo No.2 Hospital.
METHODS
The data on patients with colorectal cancer were obtained from Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2023. Patients were classified into training (n = 29017) and testing sets (n = 12437). The data were used to build eight machine learning models to predict liver metastasis in colorectal cancer patients. A total of 11 clinical variables were entered into these models. Model performance was measured with the area under the receiver operating characteristic curve (ROC) and area under precision-recall curve (AUPR). The models were visualized and interpreted using the SHAP method.
RESULTS
In the SEER database cohort, the incidence of liver metastasis was 7.2% (2977/41,454). Of the eight machine learning models, Gradient Boosting (GB) had the best AUC (0.837) and AUPR (0.294). Upon external validation, the GB model achieved an AUC of 0.730 and an AUPR of 0.278. We explored the significance of features in the model through SHAP analysis. CEA, N stage and T stage were the heavily weighted factors used by the GB. An online calculator was developed for clinical use.
CONCLUSION
The GB model demonstrates robust predictive performance for liver metastasis in CRC, validated internally and externally, and presents a potentially valuable tool for clinical decision-making.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.