使用监督的CatBoost预测甲状腺癌复发:基于SHAP的可解释AI方法
Ahmad A Hanani1, Turker Berk Donmez2, Mustafa Kutlu2
1Biomedical and Clinical Basic Skills Department, Faculty of Medicine and Health Sciences, An-Najah National University, Nablus, Palestine.
Medicine
|May 29, 2025
概括
一个CatBoost分类器准确地预测了区分良好的甲状腺癌复发,优于其他模型. 沙普利添加式解释 (SHAP) 确定了关键预测因素,如治疗反应和淋巴结状况,提高了个性化患者管理的模型解释性.
科学领域:
- 在瘤学瘤学.
- 机器学习 机器学习
- 医疗信息学 医疗信息学
背景情况:
- 在高度差异化的甲状腺癌 (WDTC) 中,复发预测具有挑战性.
- 为了更好的患者管理,需要准确和可解释的模型.
- 现有的预测模型可能缺乏足够的准确性或透明度.
研究的目的:
- 开发和评估一个监督的CatBoost分类器,用于预测WDTC复发.
- 将CatBoost模型的性能与其他组合方法进行比较.
- 使用Shapley添加式解释 (SHAP) 增强模型的解释性.
主要方法:
- 利用了383名WDTC患者的数据集,这些患者具有不同的临床和病理变量.
- 预处理的数据,处理的缺失值和编码的分类特征.
- 经过训练和测试的模型使用70:30分,评估准确性和AUC ROC.
主要成果:
- CatBoost分类器实现了97%的准确性和0.99 AUC ROC,表现优于额外树木,LightGBM和XGBoost.
- SHAP分析确定了治疗反应,风险分层和淋巴结参与作为关键预测因素.
- 当地SHAP分析显示,错误分类源于过度强调单一因素.
结论:
- 监督的CatBoost分类器为WDTC复发提供了高的预测性能和可解释性.
- 整合多种预测因素可以改善复发风险评估.
- 需要对更大的数据集进行进一步的验证,以便对甲状腺癌管理进行强大的个性化.
相关概念视频
01:21Cancer Survival Analysis
329
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
329
01:15Receiver Operating Characteristic Plot
94
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
94