{"paper":{"title":"NEURON: A Neuro-symbolic System for Grounded Clinical Explainability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A neuro-symbolic system combines medical ontology with machine learning and language models to raise prediction accuracy while generating natural-language clinical explanations.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alan Pang, Anuradha Chandrasekaran, Brady D. Lund, Dimitrios Zikos, Kewei Sha, Mutlu Mete","submitted_at":"2026-05-02T02:00:02Z","abstract_excerpt":"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 L"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"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).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the RAG-grounded LLM layer produces explanations that remain clinically accurate and faithful to the underlying model when the retrieved documents or patient notes contain incomplete or conflicting information.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NEURON raises AUC from 0.74-0.77 to 0.84-0.88 on MIMIC-IV heart-failure mortality prediction while lifting human-aligned explanation scores from 0.50 to 0.85 by grounding SHAP values in SNOMED CT and patient notes via RAG-LLM.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A neuro-symbolic system combines medical ontology with machine learning and language models to raise prediction accuracy while generating natural-language clinical explanations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"163c49f2e6c63f27ed5b052aa29c23d8a596188b13a25e7791e5b2a6cca1a7dc"},"source":{"id":"2605.01189","kind":"arxiv","version":2},"verdict":{"id":"aa5f0a3d-99d5-4bd8-b447-d827826cb1a1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-09T15:22:12.223607Z","strongest_claim":"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).","one_line_summary":"NEURON raises AUC from 0.74-0.77 to 0.84-0.88 on MIMIC-IV heart-failure mortality prediction while lifting human-aligned explanation scores from 0.50 to 0.85 by grounding SHAP values in SNOMED CT and patient notes via RAG-LLM.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the RAG-grounded LLM layer produces explanations that remain clinically accurate and faithful to the underlying model when the retrieved documents or patient notes contain incomplete or conflicting information.","pith_extraction_headline":"A neuro-symbolic system combines medical ontology with machine learning and language models to raise prediction accuracy while generating natural-language clinical explanations."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.01189/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T18:36:58.107895Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:30:47.412561Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"31f13e2d28fa0c322a39e4ae9151dad08cd02ef5f44db3ffbab0aa9f91502d47"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}