BrainIAC, a new foundation model from Mass General Brigham, outperforms traditional AI approaches in analyzing brain MRI for tasks like brain age estimation and cancer prognosis.
Key Details
- 1BrainIAC was pretrained on nearly 49,000 brain MRI scans using self-supervised learning.
- 2Validated across seven diverse clinical MRI tasks, including brain age, dementia risk, tumor mutation detection, and survival prediction.
- 3Outperformed three conventional, task-specific AI models, especially in scenarios with limited labeled data.
- 4Demonstrated strong generalizability across healthy and diseased cases as well as different MRI types.
- 5Study published in Nature Neuroscience, developed by Mass General Brigham, and funded by NIH/NCI among others.
Why It Matters
This work introduces a flexible, generalizable AI tool that could streamline the adoption of imaging AI in clinical practice, particularly where annotated data is limited. BrainIAC's ability to perform well across a range of complex imaging tasks may help accelerate personalized medicine and clinical decision making in neuroradiology.
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.