Objective Diagnosis of Fibromyalgia Using Neuroretinal Evaluation and Artificial Intelligence

Area of Science:

• Ophthalmology
• Neuroscience
• Medical Imaging

Background:

• Fibromyalgia (FM) diagnosis relies on subjective criteria.
• Objective biomarkers are needed for accurate and early FM detection.
• Neuroretinal structural changes may serve as objective indicators for FM.

Purpose of the Study:

• To identify objective biomarkers for fibromyalgia using artificial intelligence.
• To analyze structural data of the neuroretina with swept-source optical coherence tomography (SS-OCT).
• To develop a non-invasive diagnostic aid for FM.

Main Methods:

• Collected SS-OCT data on retinal and choroidal thicknesses in 29 FM patients and 32 controls.
• Analyzed thicknesses of the complete retina, ganglion cell layers (GCL+, GCL++), and retinal nerve fiber layer (RNFL).
• Employed artificial intelligence (ensemble RUSBoosted tree classifier) and AUC analysis for discriminant capacity.

Main Results:

• Significant differences in RNFL thickness were observed in the inner inferior region (p=0.010).
• Significant differences in GCL+, GCL++, and complete retina were found in four inner ring regions.
• The AI classifier achieved 82% accuracy and an AUC of 0.82.

Conclusions:

• Retinal layer analysis using SS-OCT shows potential for objective FM biomarker identification.
• This non-invasive method offers a novel approach to FM diagnosis.
• AI-driven analysis of neuroretinal structure could aid in FM detection.