Publikation
Development of an artificial intelligence-generated, explainable treatment recommendation system for urothelial carcinoma and renal cell carcinoma to support multidisciplinary cancer conferences
Gregor Duwe; Dominique Mercier; Verena Kauth; Kerstin Moench; Vikas Rajashekar; Markus Junker; Andreas Dengel; Axel Haferkamp; Thomas Hoefner
In: medrxiv, Vol. Februar, Pages 1-30, medrxiv, 2/2025.
Zusammenfassung
Background and Objective: Decisions on the best available treatment in clinical oncology are based on expert opinions in multidisciplinary cancer conferences (MCC). Artificial intelligence (AI) could increase evidence-based treatment by generating additional treatment recommendations (TR). We aimed to develop such an AI system for urothelial carcinoma (UC) and renal cell carcinoma (RCC). Methods: Comprehensive data of patients with histologically confirmed UC and RCC who received MCC recommendations in the years 2015 to 2022 were transformed into machine readable representations. Development of a two-step process to train a classifier to mimic TR. Identification of superordinate categories of recommendations followed by specification of detailed TR. Machine learning (CatBoost, XGBoost, Random Forest) and deep learning (TabPFN, TabNet, SoftOrdering CNN, FCN) techniques were trained. Results were measured by F1-scores for accuracy weights. Additionally, clinical trial data for drugs were included. Key Findings and Limitations: AI training was performed with 1617 (UC) and 880 (RCC) MCC recommendations (77 and 76 patient input parameters). AI system generated fully automated TR with excellent F1-scores for UC (e.g. Surgery 0.81, Anti-cancer drug 0.83, Gemcitabine/Cisplatin 0.88) and RCC (e.g. Anti-cancer drug 0.92 Nivolumab 0.78, Pembrolizumab/Axitinib 0.89). Explainability is provided by clinical features and their importance score. TR and explainability were visualized on a dashboard. Main limitations: single-centre and retrospective study. Conclusions and Clinical Implications: First AI-generated explainable TR in UC and RCC with excellent performance results. Potential support tool for high-quality, evidence-based TR in MCC. Study sets global reference standards for AI development in MCC recommendations in clinical oncology.