Jin Heng, Morgan Jessica I W, Gee James C, Chen Min
School of Automation Science and Electrical Engineering, Beihang University, China.
Department of Radiology, University of Pennsylvania, USA.
Proc IEEE Int Symp Biomed Imaging. 2020 Apr;2020:1383-1386. doi: 10.1109/isbi45749.2020.9098455. Epub 2020 May 22.
Adaptive optics (AO) scanning laser ophthalmoscopy offers cellular level in-vivo imaging of the human cone mosaic. Existing analysis of cone photoreceptor density in AO images require accurate identification of cone cells, which is a time and labor-intensive task. Recently, several methods have been introduced for automated cone detection in AO retinal images using convolutional neural networks (CNN). However, these approaches have been limited in their ability to correctly identify cones when applied to AO images originating from different locations in the retina, due to changes to the reflectance and arrangement of the cone mosaics with eccentricity. To address these limitations, we present an adapted CNN architecture that incorporates spatial information directly into the network. Our approach, inspired by conditional generative adversarial networks, embeds the retina location from which each AO image was acquired as part of the training. Using manual cone identification as ground truth, our evaluation shows general improvement over existing approaches when detecting cones in the middle and periphery regions of the retina, but decreased performance near the fovea.
自适应光学（AO）扫描激光检眼镜可提供人视锥细胞镶嵌的细胞水平体内成像。对AO图像中视锥光感受器密度的现有分析需要准确识别视锥细胞，这是一项耗时且费力的任务。最近，已经引入了几种使用卷积神经网络（CNN）在AO视网膜图像中自动检测视锥细胞的方法。然而，由于视锥细胞镶嵌的反射率和排列随偏心率而变化，这些方法在应用于源自视网膜不同位置的AO图像时，正确识别视锥细胞的能力有限。为了解决这些限制，我们提出了一种经过改进的CNN架构，该架构将空间信息直接纳入网络。我们的方法受条件生成对抗网络的启发，将获取每个AO图像的视网膜位置作为训练的一部分进行嵌入。以手动视锥细胞识别作为基准事实，我们的评估表明，在检测视网膜中部和周边区域的视锥细胞时，与现有方法相比有总体改进，但在中央凹附近性能下降。
