The longer a person has diabetes, the more likely they will develop DR. Among them, 20 to 30% will experience visual loss, and DR is the main cause for legal blindness for people ages 20 to 65.
Diabetes patients currently go to nearby clinics for regular checkups. Based on the physicians diagnosis, if necessary, patients will be referred to an ophthalmologist for a fundus examination, but s/he will need to wait before knowing the result. Because of the time, cost, and effort need for another exam, this causes a low DR screening rate.
Our targets are clinics and health care center with color fundus cameras. Integration of VeriSee DR into their original diagnostics process can greatly assist physicians and save time with DR screenings. The VeriSee DR generated report will show whether the patient should be recommended for referral for further examination or not.
VeriSee is an AI-assisted diagnostics software that learns from the experiences of the many ophthalmologists from the National Taiwan University Hospital in diagnosing diabetic retinopathy (DR). We use AI and deep learning techniques to make a software that achieves a level of diagnostic confidence similar to that of an ophthalmologist. Through Acer's design experience in the PC industry, this software and hardware are then integrated into an edge computing device, allowing it to compute locally without having to go through a network. We believe that this device will be of tremendous assistance for physicians in diagnosing DR.
After installing the VeriSee DR onto a laptop connected to the color fundus camera, set up the captured image exporting folder path through the options menu. When you take a fundus image with the color fundus camera, VeriSee DR will automatically begin guiding you through the analysis process.
VeriSee DR also supports a report editing mode that allows users to edit fields as necessary, adding them into the PDF report.
VeriSee DR batch mode allows the analysis of up to 1,000 fundus images, and outputting all results into a CSV file.