Article

More Harm than Help? Evaluating the Capabilities of Vision-Language Models in Neurological Image Analysis

2025-06-16

Abstract excerpt

<title>Abstract</title> <p>Objectives: This study evaluates the performance of both open-source and commercial Vision Language Models (VLMs) in interpreting radiological images of neurological diseases, comparing their diagnostic accuracy to that of experienced neuroradiologists. <h4>Methods:</h4> A dataset comprising 100 cases of brain and spine pathologies with confirmed diagnoses was curated from the Radiopae...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 3ae6ad32-c060-5c29-b541-f73c2560f248
DOI: 10.21203/rs.3.rs-6183659/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

More Harm than Help? Evaluating the Capabilities of Vision-Language Models in Neurological Image AnalysisDOI 10.21203/rs.3.rs-6183659/v1Read

Prompt Engineering Strategies Improve the Diagnostic Accuracy of GPT-4 Turbo in Neuroradiology CasesSemantic proximity Read

The Expertise Paradox: Who Benefits from LLM-Assisted Brain MRI Differential Diagnosis?Semantic proximity Read

Performance of Open-Source LLMs in Challenging Radiological Cases – A Benchmark Study on 1,933 Eurorad Case ReportsSemantic proximity Read

Diagnostic Performance of Large Language Models and Radiologists in Case-Based Radiology QuestionsSemantic proximity Read

Impact of Multimodal Prompt Elements on Diagnostic Performance of GPT-4(V) in Challenging Brain MRI CasesSemantic proximity Read

Large Language Models in Radiology Reporting—A Systematic Review of Performance, Limitations, and Clinical ImplicationsSemantic proximity Read

Learning neuroimaging models from health system-scale dataSemantic proximity Read

Comparative Analysis of GPT-4Vision, GPT-4 and Open Source LLMs in Clinical Diagnostic Accuracy: A Benchmark Against Human ExpertiseSemantic proximity Read

Board-Level Performance of Leading Open-Weight Vision-Language Models on the Japanese Diagnostic Radiology Board Examination: Reasoning, Image-Input, and Language EffectsSemantic proximity Read

Diagnostic Performances of GPT-4o, Claude 3 Opus, and Gemini 1.5 Pro in “Diagnosis Please” CasesSemantic proximity Read

Limitations in Chest X‐Ray Interpretation by Vision‐Capable Large Language Models, Gemini 1.0, Gemini 1.5 Pro, GPT‐4 Turbo, and GPT‐4oSemantic proximity Read

Comparative Analysis of ChatGPT’s Diagnostic Performance with Radiologists Using Real-World Radiology Reports of Brain TumorsSemantic proximity Read