Chen Siyi, Hoang Linh, Kashani Amir H, Yi Ji
Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD 21231, USA.
Department of Ophthalmology, Johns Hopkins University, School of Medicine, Baltimore, MD 21231, USA.
Sci Adv. 2025 Apr 4;11(14):eado8268. doi: 10.1126/sciadv.ado8268.
Vasculature morphology and hierarchy are essential for blood perfusion. Human retinal circulation is an intricate vascular system emerging and remerging at the optic nerve head (ONH). Tracing retinal vascular branching from ONH can allow detailed morphological quantification, and yet remains a challenging task. We presented a robust semi-automatic vessel tracing algorithm on human fundus images by an instance segmentation neural network (InSegNN). InSegNN separates and labels individual vascular trees and enables tracing each tree throughout its branching. We have three strategies to improve robustness and accuracy: pseudotemporal learning, spatial multisampling, and dynamic probability map. We achieved 83% specificity, 50% improvement in symmetric best dice (SBD) compared to literature, and outperformed baseline U-net, and achieved 91% precision with 71% sensitivity. We have demonstrated tracing individual vessel trees from fundus images, and simultaneously retain vessel hierarchy information. InSegNN paves a way for subsequent analysis of vascular morphology in relation to retinal diseases.
血管形态和层级结构对于血液灌注至关重要。人体视网膜循环是一个在视神经乳头（ONH）处出现并重新汇合的复杂血管系统。从视神经乳头追踪视网膜血管分支可以进行详细的形态学量化，但仍然是一项具有挑战性的任务。我们通过实例分割神经网络（InSegNN）提出了一种用于人体眼底图像的强大半自动血管追踪算法。InSegNN分离并标记各个血管树，并能够在其整个分支过程中追踪每棵树。我们有三种策略来提高鲁棒性和准确性：伪时间学习、空间多重采样和动态概率图。我们实现了83%的特异性，与文献相比对称最佳骰子系数（SBD）提高了50%，优于基线U-net，并以71%的灵敏度实现了91%的精度。我们已经证明了能够从眼底图像中追踪单个血管树，并同时保留血管层级信息。InSegNN为后续分析与视网膜疾病相关的血管形态铺平了道路。
