Research
Overview
The retina serves as a unique assessment window, containing both neural tissue and microvasculature that mirror systemic health status. As an extension of the central nervous system, the retina exhibits pathological changes that often precede clinical manifestations of neurodegenerative and vascular conditions.
The Biology
The retina's predictive power stems from its unique biological properties:
The retina is composed of specialized neurons and support cells that share many characteristics with brain tissue. Changes in retinal neurons often parallel those occurring in the brain, providing early signals of neurodegenerative conditions.
The retina's blood vessels are structurally similar to those in the brain and heart. Changes in retinal vessel diameter, tortuosity, and branching patterns can indicate systemic vascular health. This makes the retina an ideal proxy for assessing cardiovascular risk.
Similar to the blood-brain barrier, the blood-retina barrier is sensitive to systemic metabolic changes. Disruptions in this barrier can signal broader health issues before they become clinically apparent.
Our Technology
Florence combines state-of-the-art retinal imaging with advanced artificial intelligence to detect early signs of disease. We analyze three types of retinal images:
Optical Coherence Tomography (OCT) provides detailed cross-sectional views of the retina's layers with microscopic precision. This allows us to:
Measure the thickness of individual retinal layers
Assess neural tissue integrity
Evaluate microvascular structure
Detect subtle changes in retinal architecture
Color Fundus Photography (CFP) captures high-resolution images of the retinal surface, enabling:
Analysis of blood vessel patterns
Assessment of optic nerve health
Evaluation of overall retinal integrity
Detection of microvascular changes
Optical Coherence Tomography Angiography (OCT-A) provides highly detailed 3D mappings of retinal vasculature. Currently used primarily for research, this advanced imaging technique holds promise to:
Detect more subtle changes in blood vessel patterns associated with lifestyle or medical interventions
Identify these changes earlier than traditional imaging methods
Enhance the accuracy of predictive analyses for retinal health
What We Can Detect
Leveraging a transfomer-based model architecture and a semi-supervised learning approach, our technology infers subtle patterns from retinal images. This enables the early detection of neurodegenerative and vascular changes, ensuring timely intervention.
By detecting these conditions early, we enable intervention at the most effective time - before significant progression occurs.
Looking Forward
Through continued research and development, we're expanding our capabilities to detect an even broader range of health conditions, making every eye exam an opportunity for early intervention and better health outcomes.