Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
Key Details
- 1A study compared OpenAI's o1 LLM against physicians for clinical diagnostic reasoning across triage, first physician contact, and admission.
- 2The LLM matched or exceeded human diagnostic performance, especially at the ER triage stage with minimal information.
- 3Researchers highlight current LLMs' inability to integrate nontextual data (e.g., visuals, physical cues), limiting their solo clinical utility.
- 4Potential near-term uses include LLMs providing second opinions to detect diagnostic errors.
- 5Another feature covers 'digital fatigue' from EHRs and alerts, which contributes to clinician burnout and requires workflow changes.
Why It Matters
Understanding the strengths and limitations of AI models in clinical reasoning is critical for responsible adoption in radiology and broader diagnostics. Addressing digital fatigue is essential for clinician wellbeing and sustainable integration of digital tools, including AI, in medical workflows.
Source
EurekAlert
Related News
•EurekAlert
NYU AI Model Improves 5-Year Breast Cancer Risk Prediction via 3D Mammograms
NYU researchers developed an AI model using longitudinal 3D mammograms that outperforms single-scan and 2D-based tools in predicting 5-year breast cancer risk.
•EurekAlert
AI Framework Accelerates Aortic Aneurysm Risk Prediction from Imaging
Researchers developed BioPINN-LM, combining physics-informed neural networks and multimodal large language models to deliver fast, interpretable risk assessments for ascending thoracic aortic aneurysms.
•EurekAlert
Study Finds Patient Voices Missing in Generative AI Design for Oncology
A Flinders University-led review found patients and carers are rarely involved in shaping generative AI tools used in oncology.