A new AI model can accurately flag brain abnormalities in MRI scans, potentially streamlining triage and diagnosis for radiologists.
Key Details
- 1Developed by King's College London and published in Radiology: Artificial Intelligence.
- 2AI trained self-supervised on 60,000+ brain MRI scans paired with corresponding radiology reports.
- 3Successfully distinguishes normal from abnormal scans and identifies conditions like stroke, MS, and brain tumors.
- 4Allows search for similar cases using images or keywords (e.g., 'glioma').
- 5Model designed to speed up workflow by triaging scans and flagging abnormalities at time of imaging.
- 6A multicentre UK trial is planned for 2026 to evaluate real-world impact.
Why It Matters
This approach could significantly reduce radiology backlogs by automating triage and abnormality detection, supporting faster, more accurate diagnoses and potentially improving patient outcomes during radiologist shortages.
Source
EurekAlert
Related News
•EurekAlert
AI-Powered OCT Enables Rapid 'Optical Biopsy' for Early Endometrial Cancer Detection
A team at Washington University has developed a catheter-based 3D OCT system with AI to quickly and noninvasively detect early endometrial cancers.
•EurekAlert
AI Clinical Reasoning in Diagnostics and Digital Fatigue in Healthcare
Recent JMIR features explore large language models in clinical diagnostics and digital fatigue among healthcare professionals.
•EurekAlert
KAIST, MIT, Microsoft Develop Efficient AI Image Upsampling for Robotics
KAIST, MIT, and Microsoft have created 'Upsample Anything,' a training-free AI method to restore high-resolution visual data from compressed images with up to 16x improved GPU memory efficiency.