Chen Jimmy S, Marra Kyle V, Robles-Holmes Hailey K, Ly Kristine B, Miller Joseph, Wei Guoqin, Aguilar Edith, Bucher Felicitas, Ideguchi Yoichi, Coyner Aaron S, Ferrara Napoleone, Campbell J Peter, Friedlander Martin, Nudleman Eric
Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, San Diego, California.
Molecular Medicine, the Scripps Research Institute, San Diego, California.
Ophthalmol Sci. 2023 May 25;4(1):100338. doi: 10.1016/j.xops.2023.100338. eCollection 2024 Jan-Feb.
To develop a generative adversarial network (GAN) to segment major blood vessels from retinal flat-mount images from oxygen-induced retinopathy (OIR) and demonstrate the utility of these GAN-generated vessel segmentations in quantifying vascular tortuosity.
Development and validation of GAN.
Three datasets containing 1084, 50, and 20 flat-mount mice retina images with various stains used and ages at sacrifice acquired from previously published manuscripts.
Four graders manually segmented major blood vessels from flat-mount images of retinas from OIR mice. Pix2Pix, a high-resolution GAN, was trained on 984 pairs of raw flat-mount images and manual vessel segmentations and then tested on 100 and 50 image pairs from a held-out and external test set, respectively. GAN-generated and manual vessel segmentations were then used as an input into a previously published algorithm (iROP-Assist) to generate a vascular cumulative tortuosity index (CTI) for 20 image pairs containing mouse eyes treated with aflibercept versus control.
Mean dice coefficients were used to compare segmentation accuracy between the GAN-generated and manually annotated segmentation maps. For the image pairs treated with aflibercept versus control, mean CTIs were also calculated for both GAN-generated and manual vessel maps. Statistical significance was evaluated using Wilcoxon signed-rank tests ( ≤ 0.05 threshold for significance).
The dice coefficient for the GAN-generated versus manual vessel segmentations was 0.75 ± 0.27 and 0.77 ± 0.17 for the held-out test set and external test set, respectively. The mean CTI generated from the GAN-generated and manual vessel segmentations was 1.12 ± 0.07 versus 1.03 ± 0.02 ( = 0.003) and 1.06 ± 0.04 versus 1.01 ± 0.01 ( < 0.001), respectively, for eyes treated with aflibercept versus control, demonstrating that vascular tortuosity was rescued by aflibercept when quantified by GAN-generated and manual vessel segmentations.
GANs can be used to accurately generate vessel map segmentations from flat-mount images. These vessel maps may be used to evaluate novel metrics of vascular tortuosity in OIR, such as CTI, and have the potential to accelerate research in treatments for ischemic retinopathies.
The author(s) have no proprietary or commercial interest in any materials discussed in this article.
开发一种生成对抗网络（GAN），用于从氧诱导视网膜病变（OIR）的视网膜平铺图像中分割主要血管，并证明这些GAN生成的血管分割在量化血管迂曲度方面的实用性。
GAN的开发与验证。
从先前发表的手稿中获取的三个数据集，包含1084张、50张和20张平铺小鼠视网膜图像，这些图像使用了不同的染色方法，并且小鼠在处死时具有不同年龄。
四名评分者手动从OIR小鼠视网膜的平铺图像中分割主要血管。Pix2Pix（一种高分辨率GAN）在984对原始平铺图像和手动血管分割图像上进行训练，然后分别在来自保留测试集和外部测试集的100对和50对图像对上进行测试。然后，将GAN生成的和手动的血管分割图像作为输入，输入到先前发表的算法（iROP-Assist）中，以生成20对包含接受阿柏西普治疗的小鼠眼睛与对照的血管累积迂曲度指数（CTI）。
使用平均骰子系数来比较GAN生成的分割图与手动注释的分割图之间的分割准确性。对于接受阿柏西普治疗与对照的图像对，还计算了GAN生成的和手动血管图的平均CTI。使用Wilcoxon符号秩检验评估统计学显著性（显著性阈值≤0.05）。
对于保留测试集和外部测试集，GAN生成的与手动血管分割的骰子系数分别为0.75±0.27和0.77±0.17。对于接受阿柏西普治疗与对照的眼睛，GAN生成的和手动血管分割生成的平均CTI分别为1.12±0.07与1.03±0.02（P = 0.003）和1.06±0.04与1.01±0.01（P<0.001），这表明当通过GAN生成的和手动血管分割进行量化时，阿柏西普可改善血管迂曲度。
GAN可用于从平铺图像中准确生成血管图分割。这些血管图可用于评估OIR中血管迂曲度的新指标，如CTI，并有可能加速缺血性视网膜病变治疗的研究。
作者对本文中讨论的任何材料均无专利或商业利益。
