Yapay Zeka Hızlı Bakış & Çıkarımlar
AI InsightThis study is the first known to explore this issue and needs to be supported and developed with new investigations.
📌 3 Maddede Bu Makale
Abstract
Purpose: This study was conducted to evaluate the ability of multicolor (MC)–green re flectance (GR) imaging in detecting retinal non- perfused areas (RNPAs) of eyes with diabetic retinopathy (DR). Materials and Methods: Fifty-five degrees fundus fluorescein angiography (FA) and MC imaging were performed in eyes with DR. Images were divided into four fields as macular, nasal, superior temporal, and inferior temporal. To determine the presence and size of RNPAs, FA images were examined by a retina specialist and MC–GR images were examined independently by two masked retina specialists. The compliance of RNPAs in FA and MC–GR images and the agreement of retina specialists were analyzed. Results: The FA and MC–GR images of the 178 eyes were analyzed. When GR images were compared to FA images in terms of RNPAs, sensitivity was 87%; specificity was 89%; the positive predictive value was 92%; the negative predictive value was 83%; and the accuracy was 87%. A substantial agreement was observed between two retina specialists in all retinal fields. Conclusion: This study showed that 55º MC–GR imaging could be an alternative to FA in the evaluation of RNPAs. This study is the first known to explore this issue and needs to be supported and developed with new investigations.