Related Experiment Video
Updated: Jul 2, 2025
09:06
Whole-brain Segmentation and Change-point Analysis of Anatomical Brain MRI—Application in Premanifest Huntington's Disease
Published on: June 9, 2018
12.2K
Deep learning-based BMI inference from structural brain MRI reflects brain alterations following lifestyle
Ofek Finkelstein1, Gidon Levakov1, Alon Kaplan2,3
1Department of Cognitive and Brain Sciences, Ben-Gurion University of the Negev, Beer Sheva, Israel.
Human Brain Mapping
|February 20, 2024
Summary
Artificial Intelligence (AI) identified brain changes linked to weight loss in an obesity study. Predicted Body-Mass Index (BMI) reduction correlated with actual weight loss, highlighting AI
Area of Science:
- Neuroscience
- Artificial Intelligence
- Medical Imaging
Background:
- Obesity negatively impacts brain structure and function.
- Lifestyle interventions aim to mitigate obesity-related health issues.
- Understanding brain morphology changes during weight loss is crucial.
Purpose of the Study:
- To investigate if AI can detect brain morphology differences related to lifestyle interventions in overweight individuals.
- To explore the potential of AI-driven neural biomarkers for obesity and weight loss.
- To analyze brain changes associated with metabolic syndrome interventions.
Main Methods:
- Utilized an ensemble learning framework with structural brain MRIs from the DIRECT-PLUS clinical trial.
- Predicted Body-Mass Index (BMI) scores from brain MRIs before and after an 18-month lifestyle intervention.
- Employed explainable AI (XAI) to identify brain regions contributing to BMI predictions.
Main Results:
- Patient-specific BMI predictions correlated significantly with actual weight loss.
- Reductions in predicted BMI were more pronounced in active diet groups versus the control group.
- Explainable AI maps revealed distinct brain regions for BMI prediction compared to age prediction.
Conclusions:
- Predicted BMI and its reduction serve as unique neural biomarkers for obesity-related brain modifications.
- AI analysis of brain morphology can reflect clinical outcomes of lifestyle interventions.
- This approach offers novel insights into the neural underpinnings of weight loss and obesity.
