Stroke subtypes risk prediction and detection using retinal vascular structure and oxygen saturation analysis

Area of Science:

• Ophthalmology
• Neurology
• Biomarkers

Background:

• Stroke poses a significant health burden, necessitating effective screening tools for severity assessment and biomarker identification.
• Retinal biomarkers are crucial for early stroke risk detection and prognosis, enabling personalized interventions.
• This study investigates retinal blood vessel oxygen saturation (SO2) and structure in ischemic and hemorrhagic stroke subtypes versus controls.

Purpose of the Study:

• To analyze retinal blood vessel oxygen saturation (SO2) and structure across ischemic and hemorrhagic stroke subtypes compared to healthy controls.
• To examine the relationship between retinal vascular features and stroke subtypes using odds ratios.
• To classify stroke occurrence based on retinal vascular biomarkers.

Main Methods:

• Retinal images from 29 ischemic stroke patients, 23 hemorrhagic stroke patients, and 82 controls were assessed.
• Retinal blood vessel oxygen saturation (SO2) and structural parameters were analyzed using statistical and deep learning techniques.
• Logistic regression and random forest classification were employed to identify relationships and classify stroke.

Main Results:

• Ischemic stroke patients showed higher arterial SO2 than controls; hemorrhagic patients did not.
• Both stroke groups exhibited reduced arterial density and fractal dimensions compared to controls.
• Integrating SO2 and structural biomarkers achieved over 80% accuracy in classifying stroke from retinal images.

Conclusions:

• Significant differences in retinal blood vessel characteristics exist between stroke subtypes and controls.
• Retinal biomarkers are robustly associated with stroke incidence, aiding in risk anticipation.
• Retinal vascular biomarkers show potential for stroke risk stratification and cerebrovascular health applications.