{"slug": "a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential", "title": "A safety-oriented hypothetico-deductive framework for AI-assisted differential diagnosis", "summary": "Researchers developed AegisDx, a safety-oriented AI framework for differential diagnosis that coordinates specialized LLM components to generate broad diagnoses, screen for dangerous conditions, and verify reasoning against medical evidence. In evaluations on case reports and real-world emergency department notes, AegisDx improved diagnostic accuracy and safety scores compared to standalone LLMs, suggesting that engineering diagnostic AI as a reasoning framework can enhance clinical decision support.", "body_md": "arXiv:2607.08038v1 Announce Type: new\nAbstract: Diagnostic error is a major threat to patient safety, yet current large language model (LLM) systems often treat diagnosis as a one-shot prediction task, lacking safeguards against missed high-risk alternatives or rigorous verification of their reasoning. Here, we present AegisDx, a safety-oriented framework for hypothetico-deductive clinical reasoning. AegisDx coordinates specialized LLM components through role-specific contracts, structured intermediate outputs, evidence-retrieval interfaces, and verification gates to generate broad differential diagnoses, enforce explicit screening for dangerous \"must-not-miss\" conditions, verify reasoning against grounded medical evidence, and structure actionable next steps. We evaluated AegisDx across three layers. On literature-derived case reports from NEJM and JAMA, with GPT-oss-120B as the shared backbone, Top-3 diagnostic accuracy was 59.9% versus 52.1% for the standalone LLM on JAMA cases and 62.7% versus 51.4% on NEJM cases. On cases from Annals of Emergency Medicine, Top-3 accuracy was 85.7% versus 68.6%; against physician-consensus must-not-miss diagnosis sets, AegisDx captured at least one such condition among its top three diagnoses in 78.0% of cases versus 52.0%. In a blinded physician evaluation of 43 real-world emergency department notes from the Yale New Haven Health System compared against GPT-5, AegisDx improved the physician-rated composite safety score from 4.31 to 4.55 on a 5-point scale (adjusted p = 2.1x10^-4), with qualitative gains in must-not-miss identification and reasoning safety. Our findings suggest that engineering diagnostic AI as a safety-oriented reasoning framework, rather than optimizing raw predictive accuracy alone, can provide a safer, more transparent, and clinically meaningful layer of bedside decision support for acute care workflows.", "url": "https://wpnews.pro/news/a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential", "canonical_source": "https://arxiv.org/abs/2607.08038", "published_at": "2026-07-10 04:00:00+00:00", "updated_at": "2026-07-10 04:11:10.659539+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-safety", "ai-products"], "entities": ["AegisDx", "GPT-oss-120B", "NEJM", "JAMA", "Annals of Emergency Medicine", "Yale New Haven Health System", "GPT-5"], "alternates": {"html": "https://wpnews.pro/news/a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential", "markdown": "https://wpnews.pro/news/a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential.md", "text": "https://wpnews.pro/news/a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential.txt", "jsonld": "https://wpnews.pro/news/a-safety-oriented-hypothetico-deductive-framework-for-ai-assisted-differential.jsonld"}}