Yu Timothy T, Ma Da, Lo Julian, Ju Myeong Jin, Beg Mirza Faisal, Sarunic Marinko V
Engineering Science, Simon Fraser University, Burnaby BC V5A1S6, Canada.
Dept. of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, BC, V5Z 3N9, Canada.
Biomed Opt Express. 2021 Oct 1;12(10):6660-6673. doi: 10.1364/BOE.431992.
Optical coherence tomography (OCT) and OCT angiography (OCT-A) may benefit the screening of diabetic retinopathy (DR). This study investigated the effect of laterally subsampling OCT/OCT-A scans by up to a factor of 8 when using deep neural networks for automated referable DR classification. There was no significant difference in the classification performance across all evaluation metrics when subsampling up to a factor of 3, and only minimal differences up to a factor of 8. Our findings suggest that OCT/OCT-A can reduce the number of samples (and hence the acquisition time) for a volume for a given field of view on the retina that is acquired for rDR classification.
光学相干断层扫描（OCT）和OCT血管造影（OCT-A）可能有助于糖尿病视网膜病变（DR）的筛查。本研究调查了在使用深度神经网络进行自动可转诊DR分类时，将OCT/OCT-A扫描横向采样多达8倍的效果。当采样倍数高达3倍时，所有评估指标的分类性能均无显著差异，采样倍数高达8倍时也只有极小差异。我们的研究结果表明，对于为rDR分类而获取的视网膜给定视野的一个容积，OCT/OCT-A可以减少样本数量（从而减少采集时间）。
