Assessing AI and Neurologist Diagnostic Reasoning Against Neuropathological Ground Truth
Authors:
- Yu LengYLYu Leng
- Ayush NooriAyush Noori
- John R. DicksonJDJohn R. Dickson
- Alberto Serrano-PozoASAlberto Serrano-Pozo
- Eric S. RosenbergEREric S. Rosenberg
- Yingnan HeYHYingnan He
- Derek H. OakleyDODerek H. Oakley
- Vikram S. KhuranaVKVikram S. Khurana
- Bradley T. HymanBHBradley T. Hyman
- Matthew P. FroschMFMatthew P. Frosch
Preprint
Abstract
BACKGROUND Accurate differential diagnosis of complex neurological disorders remains challenging due to overlapping clinical features and heterogeneous disease presentations. Although large language models (LLMs) show promise in clinical reasoning, prior studies benchmark performance against clinician consensus rather than biological ground truth. A neuropathologically confirmed benchmark dataset for evaluating diagnostic AI in neurology is currently lacking. METHODS We introduce NeuroBench, a curated benchmark of complex neurological cases with neuropathologically confirmed gold-standard diagnoses, and DIAGNO, a confidence-aware LLM-based system for neurological diagnosis. NeuroBench comprises 203 retrospective case summaries from the Massachusetts General Hospital Brain Cutting Conference with corresponding autopsy-confirmed diagnoses. DIAGNO generated top-3 differential diagnoses, employing retrieval-augmented generation (RAG) for lower-confidence cases. Performance was assessed by three independent blinded adjudicators who evaluated both DIAGNO and neurologists against neuropathological ground truth. RESULTS NeuroBench encompassed 79 unique neuropathological diagnoses, spanning conditions including cerebrovascular disease, brain tumors, neurological infections, and various neurodegenerative and inflammatory disorders. DIAGNO matched or outperformed neurologists in top-3 accuracy (0.67 versus 0.63) and taxonomy-level accuracy (0.74 versus 0.66). In cases of disagreement, DIAGNO was more often correct than neurologists (29 versus 19 cases). Diagnostic concordance between DIAGNO and neurologists was high (90% agreement in top-3 predictions), even when both were incorrect, suggesting strong alignment in diagnostic reasoning. On NeuroBench, DIAGNO also outperformed GPT-4o baseline and DeepSeek R1 across all top-k accuracy metrics. In a real-world evaluation on eight complex cases with differentials from Mass General Brigham, neurologists rated DIAGNO's reasoning favorably (mean 4.03/5) across multiple dimensions of clinical utility and safety. CONCLUSIONS NeuroBench establishes neuropathological confirmation as the appropriate standard for evaluating diagnostic AI in neurology, moving beyond clinician-referenced benchmarking to define the ceiling of diagnostic accuracy. Evaluated against this standard, DIAGNO achieved expert-level diagnostic performance and received favorable clinician ratings in real-world applications, supporting its potential as a clinical decision-support tool in neurology.Keywords
Funding Information
- jmr barker foundation
- Bill and Rebecca Reed Endowment for Precision Therapies and Supportive Care
- Rhodes Trust
- National Institute for Health and Care Research
- NIHR Oxford Musculoskeletal Biomedical Research Centre
This publication has 29 references indexed in Scilit:
- Diagnostic reasoning prompts reveal the potential for large language model interpretability in medicinenpj Digital Medicine, 2024
- Performance of Large Language Models on a Neurology Board–Style ExaminationJAMA Network Open, 2023
- Evaluating the performance of large language models: ChatGPT and Google Bard in generating differential diagnoses in clinicopathological conferences of neurodegenerative disordersBrain Pathology, 2023
- Autopsy studies are key to identifying dementia causeThe Lancet Healthy Longevity, 2023
- Concordance of Clinical Alzheimer Diagnosis and Neuropathological Features at AutopsyJournal of Neuropathology & Experimental Neurology, 2020
- The Differential Diagnosis and Treatment of Atypical ParkinsonismDeutsches Ärzteblatt international, 2016
- Behavioral and Psychological Symptoms of DementiaFrontiers in Neurology, 2012
- National Institute on Aging–Alzheimer's Association guidelines for the neuropathologic assessment of Alzheimer's diseaseAlzheimer's & Dementia, 2012
- Consortium to Establish a Registry for Alzheimer's Disease (CERAD): The first twenty yearsAlzheimer's & Dementia, 2008
- Cognitive-Behavioral Profiles of Neurodegenerative Dementias: Beyond Alzheimer's DiseaseJournal of Geriatric Psychiatry and Neurology, 2007