Unmet Need
According to the National Institute of Health, neovascular (wet) age-related macular degeneration (nAMD) is the most common cause of visual impairment among elderly patients in developed countries, and is predicted to increase globally from 196 million patients in 2020 to 288 million by 2040. Our current management of nAMD has two major limitations. First, we are unable to identify patients at a high risk for imminent conversion from dry to nAMD, which frequently leads to delayed diagnosis and initiation of treatment. Secondly, we are unable to predict treatment response to intravitreal injections, prior to initiation of treatment , thus limiting our ability in selecting the optimal medication for each individual patient a priori.
Value Proposition:
· Predict imminent conversion to nAMD within 6 months, a clinically meaningful time window, to allow for more intense monitoring of at-risk patients.
· Predict anatomic and functional changes to anti-VEGF injections, prior to treatment initiation, to enable optimal medication selection for each patient from the beginning.
Technology Description
Deep learning based algorithms, trained with proprietary multi-modal data (including optical coherence tomography images) obtained at Johns Hopkins Medicine.
Stage of Development
A proof-of-concept system has been developed. Inventors are currently looking to perform external validation on the technology.
Data Availability
Data available upon request.
