Researchers developed a dual-branch graph attention network that uses structural and functional MRI data to accurately predict individual responses to electroconvulsive therapy in adolescents with major depressive disorder.
Key Details
- 1Study by Chongqing Medical University used deep learning (DBGAN) to fuse sMRI and fMRI data.
- 2Model tested on 27 adolescent MDD patients, achieving 85.3% accuracy and a 0.905 F1 score.
- 3Outperformed SVM, Random Forest, and standard CNNs on prediction performance.
- 4Key predictive regions: right posterior insula, dorsal cingulate gyrus (functional) and left amygdala, right hippocampus (structural).
- 5Findings suggest multi-modal MRI integration improves prediction versus single-modality approaches.
Why It Matters
The ability to predict ECT treatment response at the individual level could significantly enhance personalized psychiatry and optimize resource allocation. The integration of multi-modal imaging demonstrates the growing potential of advanced AI models in neuroimaging-driven mental health treatment.
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.