Hyperparameter-tuned prediction of somatic symptom disorder using functional near-infrared spectroscopy-based dynamic...

Area of Science:

• Neuroscience
• Medical Imaging
• Psychiatry

Background:

• Somatic symptom disorder (SSD) is characterized by medically unexplained physical symptoms causing significant distress and functional impairment.
• The neurobiological underpinnings of SSD remain largely unknown, hindering objective diagnosis.
• Current diagnosis of SSD is primarily phenomenological.

Purpose of the Study:

• To investigate the potential of functional near-infrared spectroscopy (fNIRS) in distinguishing patients with SSD from healthy controls.
• To identify neurobiological markers associated with SSD using brain functional connectivity.
• To explore the role of painful stimulation in revealing differences in brain activity between SSD patients and controls.

Main Methods:

• Utilized hyper-parameter optimized classification with Support Vector Machines (SVM).
• Employed functional near-infrared spectroscopy (fNIRS) to measure brain activity during painful stimulation (individual pain threshold and constant sub-threshold) and brush stimulation.
• Estimated dynamic functional connectivity time series using a sliding window correlation method.

Main Results:

• Achieved high classification accuracy (82%), specificity (85%), and sensitivity (81%) in distinguishing SSD patients from controls.
• Identified key brain connections, including right superior temporal-left angular gyri, right middle frontal (MFG)-left supramarginal gyri, and right middle temporal-left middle frontal gyri, as significant discriminators during the individual pain threshold condition (INDc).

Conclusions:

• fNIRS, particularly under painful stimulation conditions, shows potential for objectively distinguishing individuals with SSD from healthy controls.
• Specific brain regions, including bilateral MFG, left inferior parietal, and right temporal gyrus, may serve as robust neurobiological biomarkers for SSD.
• This study provides a foundation for further research into the neurobiology of SSD and the development of objective diagnostic tools.