Benefit

Technology Description

Researchers at the University of Iowa have developed a novel software-based diagnostic platform that leverages routine Optical Coherence Tomography (OCT) eye scans to identify biomarkers associated with coronary artery disease (CAD). Rather than relying on traditional retinal thickness measurements, the technology analyzes retinal and choroidal brightness, reflectivity patterns, and advanced image-based features extracted from OCT images. 

Using machine learning and quantitative image analysis, the platform can identify disease-associated signatures from standard ophthalmic imaging, enabling a non-invasive, rapid, and potentially scalable approach to cardiovascular risk assessment and patient stratification. This invention may also have broader applications across other systemic diseases, including stroke, neurodegenerative disorders, diabetes, and vascular diseases.

UIRF Case No. 2026-034

Stage of Development

The technology is at the research-validation stage. Its core image-analysis workflow is established: volumetric retinal OCT scans are segmented into retinal and choroidal layers, regional brightness features are quantified, and statistical and machine-learning models are used to identify CAD-associated signatures. Initial validation was completed in an angiography-verified cohort of 84 participants (60 patients with established CAD and 24 matched controls). In this cohort, lower global choroidal brightness was independently associated with CAD. A patient-level XGBoost model using bilateral OCT brightness features achieved a mean area under the receiver operating characteristic curve of 0.864, with 83% accuracy, 88% sensitivity, and 71% specificity in five-fold cross-validation. The platform has been implemented using standardized acquisition and automated analysis methods, but it has not yet undergone prospective, multi-site clinical validation. Development priorities include validating performance in larger and more diverse populations, harmonizing reflectivity measurements across OCT devices and scan protocols, refining automated quality control and reporting, and defining the clinical workflow and regulatory pathway for use as a cardiovascular risk-assessment or patient-stratification tool.