Investigators at Oregon Health & Science University (OHSU) Discuss Findings in Vision Science (High Prevalence of Artifacts In Optical Coherence Tomography With Adequate Signal Strength).

A study conducted at Oregon Health & Science University (OHSU) found a high prevalence of artifacts in optical coherence tomography (OCT) images with acceptable signal strength. The researchers analyzed 4,555 OCT images from 546 patients and identified approximately 21% of the images as nonacceptable quality. They trained and tested five convolutional neural network models to classify image quality, with the EfficientNetV2 model demonstrating superior performance. This research provides valuable insights into the challenges and potential solutions for improving OCT image quality. [Extracted from the article]