iLENS: Interpretable LLM-Guided Mixture-of-Experts for Neuroimaging Survival Analysis
For clinicians and researchers studying Alzheimer's disease, iLENS provides an interpretable survival analysis framework that bridges high-performance prediction with transparent clinical decision support.
iLENS uses LLMs to guide a mixture-of-experts model for survival prediction in Alzheimer's disease conversion, achieving competitive predictive performance and enabling patient subtyping with interpretable rationales.
Alzheimer's Disease (AD) is a complex neurodegenerative disorder that continues to impact millions of people worldwide. Predicting AD conversion during the prodromal stage remains critical for disease understanding and patient care. As such, survival models are widely used for AD risk prediction, yet they are typically static predictors with limited interpretability and no capacity for natural language reasoning. In this work, we propose iLENS, an interpretable large language model (LLM) guided framework based on mixture-of-experts (MoE) for survival prediction in AD conversion. Our approach uses LLM to synthesize structured neuroimaging measurements and unstructured information to guide expert routing. Our framework demonstrates competitive predictive performance and capability in patient subtyping. Furthermore, our framework provides transparent, biologically grounded rationales for its routing decisions, bridging the gap between high-performance survival analysis and interpretable clinical decision support.