Classification of Irritable Bowel Syndrome Using Brain Functional Connectivity Strength and Machine Learning

Area of Science:

• Neuroscience
• Medical Imaging
• Machine Learning

Background:

• Irritable Bowel Syndrome (IBS) is a common disorder linked to brain-gut interactions.
• Current understanding of IBS brain mechanisms is limited, lacking objective diagnostic tools.
• This study addresses the need for objective biomarkers in IBS diagnosis.

Purpose of the Study:

• To investigate brain network alterations in IBS using functional connectivity strength (FCS).
• To develop a machine learning classifier for distinguishing IBS patients from healthy controls.
• To identify potential fMRI-based biomarkers for IBS classification.

Main Methods:

• Resting-state fMRI scans were performed on 31 IBS patients and 30 healthy controls (HCs).
• Functional Connectivity Strength (FCS) analysis was applied to assess global brain connectivity.
• A Support Vector Machine (SVM) model was used to classify IBS patients and HCs based on FCS alterations.

Main Results:

• IBS patients exhibited increased FCS in the left medial orbitofrontal cortex (mOFC).
• Decreased FCS was observed in the bilateral cingulate cortex/precuneus (PCC/Pcu) and middle cingulate cortex (MCC) in IBS patients.
• The SVM classifier achieved 91.9% accuracy in differentiating IBS patients from HCs.

Conclusions:

• Altered FCS patterns in brain regions involved in pain and emotion regulation are characteristic of IBS.
• The identified FCS alterations show promise as effective biomarkers for IBS classification.
• This research contributes to understanding IBS neural mechanisms and improving clinical diagnosis.