In this study, we combined multimodal real-world data and explainable artificial intelligence (xAI) to introduce AI-derived (AID) markers for clinical decision support.
Multimodal data, including clinical records, image-derived body compositions, and mutational tumor profiles, from 15,726 patients from the West German Cancer Center of the University Hospital Essen across 38 cancer entities undergoing systemic treatment were included in the analysis.
Following the collection of multimodal pan-cancer data, each patient's risk score is predicted by deep learning and enables patient stratification. xAI then decomposes the patient risk into the individual contributions of each marker. This enables treatment guidance at the patient and cohort level.
The xAI model determined the prognostic contribution of each clinical marker at the patient level and identified 114 key markers that accounted for 90% of the neural network's decision process.
Moreover, the xAI model enabled us to uncover 1,373 prognostic interactions between markers.
On the other hand, performance of the xAI model compared to conventional prognostic scores showed that the xAI model significantly outperformed traditional methods in UICC Staging (C-index: 0.75 vs. 0.56), ECOG Performance Status (C-index: 0.81 vs. 0.67), and Charlson Comorbidity Index (C-index: 0.75 vs. 0.63) respectively.
Performance remained consistent across different cancer types, validating the robustness of the model.
The model was further validated in an independent cohort of 3,288 patients with lung cancer from a US nationwide electronic health record-derived database.
These results show the potential of xAI to transform the assessment of clinical variables and enable personalized, data-driven cancer care.
【MORE】