Second Opinion Matters: Towards Adaptive Clinical AI via the Consensus of Expert Model Ensemble

TL;DR

This paper introduces the Consensus Mechanism, an ensemble of specialized medical expert models, to improve clinical decision-making with adaptive, robust, and cost-effective AI systems evaluated across multiple medical benchmarks.

Contribution

The paper proposes a novel ensemble framework mimicking clinical decision processes, enhancing adaptability and accuracy over single-model approaches in medical AI applications.

Findings

Achieved 61.0% accuracy on MedXpertQA, outperforming O3 and Gemini 2.5 Pro.

Improved accuracy on MedQA and MedMCQA benchmarks by 3.4% and 9.1%.

Enhanced differential diagnosis recall, precision, and top-1 accuracy.

Abstract

Despite the growing clinical adoption of large language models (LLMs), current approaches heavily rely on single model architectures. To overcome risks of obsolescence and rigid dependence on single model systems, we present a novel framework, termed the Consensus Mechanism. Mimicking clinical triage and multidisciplinary clinical decision-making, the Consensus Mechanism implements an ensemble of specialized medical expert agents enabling improved clinical decision making while maintaining robust adaptability. This architecture enables the Consensus Mechanism to be optimized for cost, latency, or performance, purely based on its interior model configuration. To rigorously evaluate the Consensus Mechanism, we employed three medical evaluation benchmarks: MedMCQA, MedQA, and MedXpertQA Text, and the differential diagnosis dataset, DDX+. On MedXpertQA, the Consensus Mechanism achieved an accuracy of 61.0% compared to 53.5% and 45.9% for OpenAI's O3 and Google's Gemini 2.5 Pro. Improvement was consistent across benchmarks with an increase in accuracy on MedQA ($\Delta\mathrm{Accuracy}_{\mathrm{consensus\text{-}O3}} = 3.4\%$) and MedMCQA ($\Delta\mathrm{Accuracy}_{\mathrm{consensus\text{-}O3}} = 9.1\%$). These accuracy gains extended to differential diagnosis generation, where our system demonstrated improved recall and precision (F1$_\mathrm{consensus}$ = 0.326 vs. F1$_{\mathrm{O3\text{-}high}}$ = 0.2886) and a higher top-1 accuracy for DDX (Top1$_\mathrm{consensus}$ = 52.0% vs. Top1$_{\mathrm{O3\text{-}high}}$ = 45.2%).