．Speaker． 

Dr. Amy Hsiao

Manager of Product Strategy and Management, Acer Healthcare

Speaker's Biography

Dr. Hsiao has over 15 years of experience in the field of biomedical engineering. She received her doctoral degree in Biomedical Engineering from the University of Michigan and has since worked in various functions, including research and development, project management, and business development. She joined Acer Healthcare's business development team in 2020 and has recently transitioned to take on a new role in the product strategy and management team.

Topic

Overview of VeriSee DR - An AI-assisted Diagnostic Software for Diabetic Retinopathy

Abstract

With the growing number of people with diabetes worldwide, the traditional practice of manual screening of diabetic retinopathy (DR) by diabetes and eye care specialists cannot scale up. While annual DR screening is recommended for all diabetic patients, only less than 50% of patients get screened even in the developed world. To address this challenge, Acer Healthcare developed VeriSee DR, an artificial intelligence (AI) assisted diagnostic software for DR, in collaboration with ophthalmologists from National Taiwan University Hospital. VeriSee DR uses AI to analyze images of the retina for signs of DR and assists healthcare professionals, especially doctors who are not eye care specialists, to interpret the images. With the implementation of VeriSee DR, after the patient's retinal images are taken, the images will be analyzed by AI using edge computing, and a binary result (non-referable vs. referable) will be output within a few seconds. The patient will receive immediate feedback on whether he/she should be referred to an eye care professional or come back for normal annual testing. Since VeriSee DR uses edge computing, no internet connection is required, and the images do not need to be uploaded to a cloud server for analysis. With such features, VeriSee DR is expected to bring real impacts to remote areas where medical resources are scarce. VeriSee DR has a screening sensitivity of 95%, specificity of 90%, and accuracy of 93%, which are comparable to ophthalmologists. It is an effective solution that can shorten the screening process, increase the volume of screenings, improve the accuracy and consistency of the results, reduce costs, and improve patient outcomes. By optimizing and expediting the screening process, patients can receive proper treatment earlier. Healthcare providers can spend more time on diabetes management and education, and the quality of diabetes care will improve over time.

Slide and Additional Information