Doheny Eye Institute, 150 N Orange Grove Blvd, Pasadena, 91103, USA.
The University of California, Los Angeles, CA, 90095, USA.
Sci Rep. 2022 Aug 26;12(1):14565. doi: 10.1038/s41598-022-18785-6.
Age-related macular degeneration (AMD) and Stargardt disease are the leading causes of blindness for the elderly and young adults respectively. Geographic atrophy (GA) of AMD and Stargardt atrophy are their end-stage outcomes. Efficient methods for segmentation and quantification of these atrophic lesions are critical for clinical research. In this study, we developed a deep convolutional neural network (CNN) with a trainable self-attended mechanism for accurate GA and Stargardt atrophy segmentation. Compared with traditional post-hoc attention mechanisms which can only visualize CNN features, our self-attended mechanism is embedded in a fully convolutional network and directly involved in training the CNN to actively attend key features for enhanced algorithm performance. We applied the self-attended CNN on the segmentation of AMD and Stargardt atrophic lesions on fundus autofluorescence (FAF) images. Compared with a preexisting regular fully convolutional network (the U-Net), our self-attended CNN achieved 10.6% higher Dice coefficient and 17% higher IoU (intersection over union) for AMD GA segmentation, and a 22% higher Dice coefficient and a 32% higher IoU for Stargardt atrophy segmentation. With longitudinal image data having over a longer time, the developed self-attended mechanism can also be applied on the visual discovery of early AMD and Stargardt features.
年龄相关性黄斑变性 (AMD) 和斯特格德特病分别是导致老年人和年轻人失明的主要原因。AMD 的脉络膜萎缩和斯特格德特萎缩是它们的终末期结果。高效的方法来分割和量化这些萎缩病变对于临床研究至关重要。在这项研究中，我们开发了一种带有可训练的自我注意机制的深度卷积神经网络 (CNN)，用于准确分割 GA 和 Stargardt 萎缩。与传统的后处理注意机制只能可视化 CNN 特征不同，我们的自我注意机制被嵌入到全卷积网络中，并直接参与训练 CNN 来主动关注关键特征，从而提高算法性能。我们将自我注意 CNN 应用于眼底自发荧光 (FAF) 图像上的 AMD 和 Stargardt 萎缩病变的分割。与现有的常规全卷积网络 (U-Net) 相比，我们的自我注意 CNN 在 AMD GA 分割方面的 Dice 系数提高了 10.6%，IoU（交并比）提高了 17%，在 Stargardt 萎缩分割方面的 Dice 系数提高了 22%，IoU 提高了 32%。随着时间的推移，有更多的纵向图像数据，所开发的自我注意机制也可以应用于早期 AMD 和 Stargardt 特征的视觉发现。
