从像素到预后:使用SHAP和LIME揭示放射学模型,以提高可解释性
Sotiris Raptis1, Christos Ilioudis2, Kiriaki Theodorou1
1Medical Physics Department, Medical School, University of Thessaly, Larisa 41500, Greece.
Biomedical physics & engineering express
|March 18, 2024
概括
这项研究引入了使用多模式成像数据预测疾病进展的先进放射学模型. 深度神经网络表现出卓越的性能,通过SHAP和LIME等可解释性方法来提高临床信任.
科学领域:
- 放射学和医学成像学 医学成像学
- 人工智能在医学中的应用
- 计算病理学计算病理学
背景情况:
- 放射性肺炎 (RP) 是胸部放射治疗的常见并发症.
- 现有的放射学模型主要专注于预测RP,对更广泛的疾病进展的范围有限.
- 需要先进的预测模型,整合多模式数据,以获得全面的预测见解.
研究的目的:
- 开发和评估基于多模式放射学的预测模型,用于超越辐射肺炎的疾病进展.
- 为了比较渐变增强机器 (GBM) 与深度神经网络 (DNN) 的性能.
- 为了提高模型的解释性,使用SHAP和LIME进行临床应用.
主要方法:
- 使用PyRadiomics从PET/CT成像中提取放射性特征 (强度,纹理,形状).
- 预测模型的开发,包括GBM变体 (XGBoost,LightGBM,CatBoost) 和DNNs.
- 使用多模式AUC-ROC,灵敏度,特异性和F1-Score进行性能评估,并进行外部验证.
主要成果:
- 深度神经网络 (DNN) 模型实现了0.90.9的多模式AUC-ROC.
- DNN 显示出高灵敏度 (0.85) 和特异性 (0.91).
- SHAP和LIME方法提供了本地化的解释,提高了模型的可解释性.
结论:
- 多模式放射学模型,特别是DNN,在预测复杂疾病进展方面显示出显著的前景.
- 模型解释性技术对于临床采用和对人工智能驱动预测的信任至关重要.
- 这种方法意味着从像素级图像分析向预后见解的转变.
相关概念视频
Radiological Investigation II: MRI and Ventilation Perfusion Scan
933
Description
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
933
Imaging Studies for Cardiovascular System IV: CMRI
550
Cardiovascular magnetic resonance imaging, or CMRI, is a non-invasive diagnostic test that employs a magnetic field and radiofrequency waves to create precise images of the heart and arteries. It provides comprehensive information about cardiac anatomy, function, perfusion, and tissue characterization without ionizing radiation.IndicationsCMRI diagnoses various heart conditions, including tissue damage from heart attacks, ischemic heart disease, myocarditis, aortic issues (tears, aneurysms,...
550


