Background

Granular dystrophy is the most common stromal dystrophy. To perform automated segmentation of corneal stromal deposits, we trained and tested a deep learning (DL) algorithm from patients with corneal stromal dystrophy and compared its performance with human segmentation.

Methods

In this retrospective cross-sectional study, we included slit-lamp photographs by sclerotic scatter from patients with corneal stromal dystrophy and real-world slit-lamp photographs via various techniques (diffuse illumination, tangential illumination, and sclerotic scatter). Our data set included 1007 slit-lamp photographs of semi-automatically generated handcraft masks on granular and linear lesions from corneal stromal dystrophy patients (806 for the training set and 201 for test set). For external test (140 photographs), we applied the DL algorithm and compared between automated and human segmentation. For performance, we estimated the intersection of union (IoU), global accuracy, and boundary F1 (BF) score for segmentation.

Results

In 201 internal test set, IoU, global accuracy, and BF score with 95 % confidence Interval were 0.81 (0.79–0.82), 0.99 (0.98–0.99), and 0.93 (0.92–0.95), respectively. In 140 heterogenous external test set as a real-world data, those were 0.64 (0.61–0.67), 0.95 (0.94–0.96), and 0.70 (0.64–0.76) via DL algorithm and 0.56 (0.51–0.61), 0.95 (0.94–0.96), and 0.70 (0.65–0.74) via human rater, respectively.

Conclusions

We developed an automated segmentation DL algorithm for corneal stromal deposits in patients with corneal stromal dystrophy. Segmentation on corneal deposits was accurate via the DL algorithm in the well-controlled dataset and showed reasonable performance in a real-world setting. We suggest this automatic segmentation of corneal deposits helps to monitor the disease and can evaluate possible new treatments.

中文翻译：

---

通过深度学习从角膜基质营养不良图像中自动分割角膜沉积物

背景

颗粒性营养不良是最常见的间质营养不良。为了对角膜基质沉积进行自动分割，我们训练并测试了角膜基质营养不良患者的深度学习 (DL) 算法，并将其性能与人类分割进行了比较。

方法

在这项回顾性横断面研究中，我们包括了角膜基质营养不良患者的硬化散射裂隙灯照片和通过各种技术（漫射照明、切向照明和硬化散射）的真实裂隙灯照片。我们的数据集包括 1007 张裂隙灯照片，这些照片是半自动生成的手工艺面具，用于角膜基质营养不良患者的颗粒状和线状病变（806 张用于训练集，201 张用于测试集）。对于外部测试（140 张照片），我们应用了 DL 算法并在自动分割和人工分割之间进行了比较。对于性能，我们估计了用于分割的联合 (IoU)、全局准确度和边界 F1 (BF) 分数的交集。

结果

在 201 内部测试集中，IoU、全局准确率和 95% 置信区间的 BF 分数分别为 0.81（0.79-0.82）、0.99（0.98-0.99）和 0.93（0.92-0.95）。在作为真实世界数据的 140 个异类外部测试集中，通过 DL 算法分别为 0.64 (0.61–0.67)、0.95 (0.94–0.96) 和 0.70 (0.64–0.76) 以及 0.56 (0.51–0.65) (0.94)。 –0.96) 和 0.70 (0.65–0.74) 分别通过人工评估。

结论

我们开发了一种自动分割 DL 算法，用于角膜基质营养不良患者的角膜基质沉积。在控制良好的数据集中，通过 DL 算法对角膜沉积物进行分割是准确的，并在现实环境中表现出合理的性能。我们建议这种角膜沉积物的自动分割有助于监测疾病并评估可能的新治疗方法。