NEURON: A Neuro-symbolic System for Grounded Clinical Explainability

2026-05-01

AI Summary

This study addresses the challenge of achieving clinically meaningful interpretability in artificial intelligence due to the lack of ontological grounding and narrative transparency. To bridge this gap, the authors propose a novel neuro-symbolic approach that integrates the SNOMED CT ontology with machine learning models for the first time. By leveraging a retrieval-augmented generation (RAG)-enhanced large language model, the method translates SHAP-based feature attributions and patient records into natural language explanations aligned with human clinical reasoning. Evaluated on acute heart failure mortality prediction using the MIMIC-IV dataset, the model achieves an AUC of 0.84–0.88, while its generated explanations attain a human-alignment score of 0.85—substantially higher than the baseline SHAP explanations (0.50)—demonstrating a unified framework for both high-performance prediction and trustworthy, clinically interpretable outputs.

Abstract

Clinical AI adoption is hindered by the black-box/grey-box nature of high-performing models, which lack the ontological grounding and narrative transparency required for professional-level explainability. We present NEURON, a neuro-symbolic system designed to enhance both predictive reliability and clinical interpretability. NEURON integrates SNOMED CT ontology-informed structural representations with machine learning models to bridge the gap between raw data and medical nomenclature. To facilitate human-aligned interaction, the system utilizes a Retrieval-Augmented Generation (RAG) grounded LLM layer to synthesize SHAP feature attributions and patient-specific clinical notes into coherent, natural-language explanations. Validated on the MIMIC-IV dataset for Acute Heart Failure mortality prediction, NEURON improved the AUC from 0.74-0.77 to 0.84-0.88 and significantly outperformed raw SHAP visualizations in human-aligned metrics (0.85 vs. 0.50). Our results demonstrate that NEURON offers a robust, scalable engineering solution for deploying trustworthy, human-centered connected health applications.

Problem

• clinical explainability
• black-box models
• ontological grounding
• narrative transparency
• trustworthy AI

Innovation

• neuro-symbolic
• clinical explainability
• SNOMED CT
• Retrieval-Augmented Generation
• SHAP