Brazilian and French researchers have developed an imaging-based AI tool to predict how multiple sclerosis patients will respond to natalizumab treatment.
Key Details
- 1Combines high-content cell imaging and machine learning to analyze patient blood samples before natalizumab therapy.
- 2Study used over 400 cell morphological features, with 130 key characteristics for prediction.
- 3Tool achieved 92% accuracy in discovery and 88% in validation cohorts for predicting drug response.
- 4Non-responders showed distinct actin remodeling and cell morphology (more elongated CD8+ T cells).
- 5Findings published in Nature Communications, suggesting potential for broader disease and drug applications.
Why It Matters
This research demonstrates how imaging AI and advanced cell analysis can enable more personalized and effective treatment strategies in neurology. Such approaches could optimize resource use, reduce side effects, and accelerate precision medicine adoption in radiology and adjacent fields.
Source
EurekAlert
Related News
•EurekAlert
AI-Based Imaging Biomarker for Chronic Stress Validated on Chest CT
A deep learning model has identified and validated the first imaging biomarker for chronic stress using adrenal gland measurements from routine chest CT scans.
•EurekAlert
UK Researchers Launch Unbiased AI Testing Platform for Diabetic Retinopathy Screening
Researchers unveiled a new independent platform for fair, large-scale evaluation of commercial AI algorithms detecting diabetic eye disease within the NHS.
•EurekAlert
Multi-Algorithm Machine Learning Model Enhances Exacerbation Risk Prediction in COPD
A multi-algorithm machine learning model significantly improves risk prediction of acute exacerbations in COPD patients using multidimensional clinical data.