Domain-Expert-Guided Hybrid Mixture-of-Experts for Medical AI: Integrating Data-Driven Learning with Clinical Priors

TL;DR

This paper introduces DKGH-MoE, a hybrid model that combines data-driven learning with clinical priors, such as eye-gaze cues, to enhance medical AI performance and interpretability.

Contribution

It presents a novel plug-and-play module that unifies data-driven and expert-guided experts within a mixture-of-experts framework for medical imaging.

Findings

Improved diagnostic accuracy over baseline models

Enhanced interpretability through clinical priors

Effective integration of domain knowledge with data-driven features

Abstract

Mixture-of-Experts (MoE) models increase representational capacity with modest computational cost, but their effectiveness in specialized domains such as medicine is limited by small datasets. In contrast, clinical practice offers rich expert knowledge, such as physician gaze patterns and diagnostic heuristics, that models cannot reliably learn from limited data. Combining data-driven experts, which capture novel patterns, with domain-expert-guided experts, which encode accumulated clinical insights, provides complementary strengths for robust and clinically meaningful learning. To this end, we propose Domain-Knowledge-Guided Hybrid MoE (DKGH-MoE), a plug-and-play and interpretable module that unifies data-driven learning with domain expertise. DKGH-MoE integrates a data-driven MoE to extract novel features from raw imaging data, and a domain-expert-guided MoE incorporates clinical priors, specifically clinician eye-gaze cues, to emphasize regions of high diagnostic relevance. By integrating domain expert insights with data-driven features, DKGH-MoE improves both performance and interpretability.