Predicting acupuncture efficacy for functional dyspepsia based on functional brain network features: a machine...

Area of Science:

• Neuroscience
• Gastroenterology
• Medical Imaging

Background:

• Acupuncture shows efficacy in treating functional dyspepsia (FD), but patient response varies.
• Objective biomarkers are needed to predict individual responses to acupuncture for FD.

Purpose of the Study:

• To develop a predictive model for acupuncture responsiveness in FD patients using pre-treatment functional brain network data.
• To identify key brain network features that predict treatment outcomes in FD patients undergoing acupuncture.

Main Methods:

• Support vector machine models were constructed using functional brain MRI data from 100 FD patients.
• Pre-treatment functional brain networks were analyzed as predictive features.
• Longitudinal analysis compared brain network changes between responders and non-responders.

Main Results:

• The prediction model achieved 76% accuracy in distinguishing acupuncture responders from non-responders.
• The model explained 24% of the variance in dyspeptic symptom relief.
• Thirty-eight critical predictive features involving the orbitofrontal cortex, caudate, hippocampus, and anterior insula were identified.

Conclusions:

• Functional brain network analysis offers a promising method for predicting acupuncture efficacy in FD patients.
• Identifying predictive biomarkers can guide personalized acupuncture treatment strategies for FD.
• Brain network alterations are more significant in acupuncture responders, highlighting their role in treatment success.