In this SBIR project, we present EyeQuant, an AI-enabled fully automated tool to com- pute retinal image-based biomarkers for mild cognitive impairment and neurodegenera- tive disorders. The large, growing, and aging population is likely to lead to a dramatic increase in the number of people at risk for dementia and other cognitive disorders, with more than 130 million people expected to be affected by 2050. Vascular disease is an important cause of dementia and cognitive decline in older people. The retinal microvas- culature, which can be non-invasively captured on color fundus photographs, share em- bryological origins, anatomical features, and physiological properties with the cerebral microvasculature. Biomarkers capturing structural changes in retinal microvasculature have been shown to be indicative of cognitive disorders in patients, including mild cogni- tive impairment (MCI), Alzheimer's disease, and vascular dementia. These structural vas- culature changes can manifest themselves earlier than any functional changes commonly associated with these cognitive disorders. A tool, such as EyeQuant, that can automati- cally and easily characterize these structural changes as biomarkers directly in primary care settings using fundus photographs, would enable early detection and effective man- agement of patients with cognitive and neurodegenerative disorders.
Public Health Relevance Statement: Narrative EyeQuant is an AI-enabled fully automated tool to compute retinal image-based biomarkers for mild cognitive impairment and neurodegenerative disorders. With more than 130 million people expected to be effected by dementia and other cognitive disorders by 2050, the burden and global annual costs of 818 billion USD is expected to increase significantly over the coming decades. A tool, such as EyeQuant, that can automatically and easily characterize structural retinal microvasculature changes as biomarkers directly in primary care settings using non- invasively captured color fundus photographs, would enable early detection and effective management of patients with cognitive and neurodegenerative disorders.
NIH Spending Category: Acquired Cognitive Impairment; Aging; Alzheimer's Disease; Alzheimer's Disease Related Dementias (ADRD); Alzheimer's Disease including Alzheimer's Disease Related Dementias (AD/ADRD); Bioengineering; Biomedical Imaging; Brain Disorders; Cerebrovascular; Clinical Research; Dementia; Eye Disease and Disorders of Vision; Machine Learning and Artificial Intelligence; Networking and Information Technology R&D (NITRD); Neurodegenerative; Neurosciences; Prevention; Vascular Cognitive Impairment/Dementia
Project Terms: Affect; aging population; Agreement; Algorithms; Alzheimer's Disease; Anatomy; Architecture; arteriole; base; Biological Markers; Blood Vessels; Caliber; cerebral microvasculature; Characteristics; Clinical; Clinical Data; Cloud Computing; Cognition Disorders; Color; Computer software; Computer Vision Systems; Consumption; cost; Data; Data Collection; Data Set; Databases; deep learning; Dementia; dementia risk; design; Development; Diabetic Retinopathy; Diagnosis; Dimensions; Early Diagnosis; early screening; experience; Eye; Fractals; Fundus; fundus imaging; Fundus photography; Goals; Gold; Hour; Human; Image; Image Analysis; imaging biomarker; Impaired cognition; interest; Lead; Los Angeles; Machine Learning; macula; Manuals; Measures; Medical center; mild cognitive impairment; Neurodegenerative Disorders; Neurologist; Neurology; novel; operation; Ophthalmology; Patients; Pharmacotherapy; Phase; Physiological; point of care; primary care setting; Primary Health Care; Process; Property; Retinal; Retinal Diseases; retinal imaging; Scientist; screening; Screening procedure; serial imaging; Small Business Innovation Research Grant; software development; Software Engineering; Structure; success; Surveys; System; Systems Analysis; Testing; Time; tool; Treatment Efficacy; usability; user-friendly; Validation; Vascular Dementia; Vascular Diseases; venule; Work
