NEURON: A Neuro-symbolic System for Grounded Clinical Explainability

TL;DR

NEURON is a neuro-symbolic system that improves clinical AI explainability and reliability by integrating ontological knowledge with machine learning and natural language explanations, validated on heart failure data.

Contribution

The paper introduces NEURON, a novel neuro-symbolic system combining ontological representations with LLMs for enhanced clinical interpretability and predictive performance.

Findings

NEURON increased AUC from 0.74-0.77 to 0.84-0.88.

It outperformed raw SHAP visualizations in human-aligned metrics (0.85 vs. 0.50).

Validated on MIMIC-IV dataset for heart failure mortality prediction.

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.