A collaboration of researchers from Xi’an Jiaotong-Liverpool University (XJTLU) and VoxelCloud Inc. in China have presented an AI-powered medical imaging technique—DualStreamFoveaNet (DSFN)—to overcome the current imaging issues. The findings are reported in the IEEE Journal of Biomedical and Health Informatics.
Dr. Sifan Song, a Ph.D. graduate from XJTLU’s School of AI and Advanced Computing and first author of the study, says
DualStreamFoveaNet (DSFN) uses retina scans and vascular distribution information. Moreover, to precisely detect the fovea—a depression at the rear of the eye where visual acuity is highest—in complex clinical circumstances.
Dr. Sifan Song says
Dr. Song explains that the surrounding retinal tissue’s color intensity gives the fovea’s dark appearance. And indistinguishable from the retinal background, which is further obscured by retinal diseases. He emphasizes that low light conditions and non-standard fovea locations during photography further challenge accurate fovea localization.
Dr. Song discusses how the design of DSFN decreases computing costs. While maintaining excellent accuracy, making it more acceptable and inexpensive for use in clinical settings.
Higher processing rates complement low computational costs. That allows doctors to acquire diagnostic results more rapidly. As well as faster model updates and iterations, which lead to more accurate forecasts of ocular disorders,” adds Dr Song.