Functional Connectome in Prader-Willi Syndrome: Neuroimaging and AI to Assess Therapeutic Impact

Behavior Disorders, Hyperphagia, Prader Willi Syndrome

Neuroimage, Artificial Intelligence, Prader-Willi Syndrome

The goal of this observational study is to explore brain network changes and identify patterns related to hyperphagia, hormonal treatment effects, and cognitive deficits in adults with Prader-Willi Syndrome (PWS). The main questions it aims to answer are: * How are brain connectivity patterns altered in PWS patients compared to healthy and obese controls? * How do brain network changes relate to hyperphagia and the response to growth hormone therapy? Researchers will compare PWS patients to healthy and obese controls to see if there are significant differences in brain network connectivity before and after meals and growth hormone therapy. Ultimately, researchers will try to develop predictive models of treatment outcomes using AI and machine learning.

Prader-Willi Syndrome (PWS) is a rare genetic condition marked by issues like constant hunger, obesity, hormonal imbalances, and cognitive difficulties. While studies have shown changes in brain connectivity in PWS patients, a complete understanding of these changes is still lacking. This project aims to explore brain network patterns in PWS using advanced AI techniques to better understand the impact of hyperphagia, hormonal treatments, and cognitive challenges. This study will combine data from brain functional magnetic ressonance imaging (fMRI)), genetic data, hormonal levels, and clinical details; of 39 adults with PWS and 82 control participants (including 52 healthy and 30 obese controls). All participants are matched for age, sex, and BMI where applicable. fMRI data has been obtained before and after meals, and/or before and after one year of growth hormone therapy. Data quality will be assessed and summarize it using averages or percentages. Then, groups will be statistically compared to detect patterns related to hyperphagia and GH therapy. Regarding Brain Connectivity, data from patients and controls will be compared, tracking how brain networks change after meals/GH therapy and applying advanced statistical methods to control errors. Finally, using AI, models to predict treatment outcomes and brain network changes will be developed. These models will be tested and refined using different techniques to ensure reliability. With this research, expected results are to identify unique brain patterns in PWS, uncover how these relate to symptoms like hyperphagia, and develop AI models to predict treatment outcomes. Ultimately, this research aims to improve our understanding of PWS and help develop better treatments.

No related papers found yet.

Spain