[论文解读] OCTolyzer: Fully automatic toolkit for segmentation and feature extracting in optical coherence tomography and scanning laser ophthalmoscopy data

Jamie Burke, Justin Engelmann+10 more|arXiv (Cornell University)|Jul 19, 2024

Retinal Imaging and AnalysisMedicine被引用 3

一句话总结

OCTolyzer 是一个开源的全自动工具包，用于从光学相干断层扫描（OCT）和扫面激光眼底镜（SLO）数据中进行分割和提取临床有意义的特征。它利用深度学习和元数据驱动的分割方法，实现高可重复性的视网膜脉络膜测量——如脉络膜厚度、体积和血管指数，单个B-scan的OCT处理在标准CPU上耗时不足2秒。

ABSTRACT

Optical coherence tomography (OCT) and scanning laser ophthalmoscopy (SLO) of the eye has become essential to ophthalmology and the emerging field of oculomics, thus requiring a need for transparent, reproducible, and rapid analysis of this data for clinical research and the wider research community. Here, we introduce OCTolyzer, the first open-source toolkit for retinochoroidal analysis in OCT/SLO data. It features two analysis suites for OCT and SLO data, facilitating deep learning-based anatomical segmentation and feature extraction of the cross-sectional retinal and choroidal layers and en face retinal vessels. We describe OCTolyzer and evaluate the reproducibility of its OCT choroid analysis. At the population level, metrics for choroid region thickness were highly reproducible, with a mean absolute error (MAE)/Pearson correlation for macular volume choroid thickness (CT) of 6.7$μ$m/0.99, macular B-scan CT of 11.6$μ$m/0.99, and peripapillary CT of 5.0$μ$m/0.99. Macular choroid vascular index (CVI) also showed strong reproducibility, with MAE/Pearson for volume CVI yielding 0.0271/0.97 and B-scan CVI 0.0130/0.91. At the eye level, measurement noise for regional and vessel metrics was below 5% and 20% of the population's variability, respectively. Outliers were caused by poor-quality B-scans with thick choroids and invisible choroid-sclera boundary. Processing times on a laptop CPU were under three seconds for macular/peripapillary B-scans and 85 seconds for volume scans. OCTolyzer can convert OCT/SLO data into reproducible and clinically meaningful retinochoroidal features and will improve the standardisation of ocular measurements in OCT/SLO image analysis, requiring no specialised training or proprietary software to be used. OCTolyzer is freely available here: https://github.com/jaburke166/OCTolyzer.

研究动机与目标

• 解决缺乏开源、自动化处理OCT和SLO数据以生成临床有意义的视网膜脉络膜特征的管道问题。
• 提升科研和临床环境中视网膜和脉络膜图像分析的标准化水平。
• 实现在OCT和SLO图像中对脉络膜厚度、体积和血管指数进行高通量、可重复的测量。
• 通过消除人工分割并减少对专有软件的依赖，支持大规模眼科图像分析。
• 使无专业训练或软件的科研人员也能访问先进的视网膜和脉络膜指标。

提出的方法

• 利用图像元数据和深度学习，实现OCT扫描中视网膜层的自动分割。
• 采用基于U-Net的深度学习模型对OCT中的脉络膜层进行分割，训练数据来自系统性健康相关数据。
• 整合俯视SLO图像，用于解剖结构分割和视网膜血管特征提取。
• 使用EyePy以Python方式导入HEYEX查看器生成的.vol RAW文件，支持SLO和OCT数据。
• 利用OCT系统校准数据，将像素坐标转换为物理空间（微米单位）。
• 从分割区域计算各项指标，包括黄斑区和视乳头周围脉络膜厚度、体积以及脉络膜血管指数（CVI）。

Figure 1 : OCTolyzer’s pipeline. (A) OCTolyzer expects a file containing the OCT (and optional SLO) image data, and necessary OCT file metadata. (B – E) A deeper dive into OCTolyzer’s internal structure. (B – D) OCT analysis suite: analysis pipelines for when OCTolyzer detects a single/radial macula

实验结果

研究问题

• RQ1 一个完全自动化的开源管道是否能在OCT和SLO数据中实现高可重复性的脉络膜厚度和体积测量？
• RQ2 OCTolyzer在不同图像质量和解剖变异情况下的OCT扫描脉络膜层分割表现如何？
• RQ3 OCTolyzer在无需人工干预的情况下，能多大程度上可靠地从SLO图像中提取俯视视网膜血管特征？
• RQ4 该管道在不同OCT扫描仪类型和采集协议下的性能表现如何比较？
• RQ5 OCTolyzer能否支持大规模、标准化的视网膜脉络膜表型分析，用于人群研究？

主要发现

• 黄斑区脉络膜体积厚度的平均绝对误差为6.7 µm，皮尔逊相关系数和斯皮尔曼等级相关系数分别为0.9933和0.9969，表明具有高度可重复性。
• 视乳头周围脉络膜厚度的平均绝对误差为5.0 µm，皮尔逊相关系数和斯皮尔曼等级相关系数分别为0.9942和0.9940，表现出极佳的可重复性。
• 黄斑区脉络膜血管指数（CVI）的平均绝对误差为0.0271（体积CVI）和0.0130（B-scan CVI），相关系数分别为0.9669和0.9090。
• 在眼水平上，区域和血管指标的测量误差分别低于人群变异性的5%和20%，表明具有鲁棒性。
• 主要异常值与图像质量差、脉络膜较厚且脉络膜-巩膜边界不清晰的B-scan有关，凸显图像质量是关键影响因素。
• 在标准笔记本电脑CPU上，OCT处理时间对黄斑区或视乳头周围B-scan不足2秒，对体积扫描为85秒，表明计算效率高。

Figure 2 : A diagram of how measurements are made for a single macular OCT B-scan. (A) localiser SLO with line of OCT acquisition shown in green. (B) Exemplar horizontal-line OCT B-scan overlaid onto the localiser SLO. (C) OCT B-scan with retinal and choroid layer segmentations and layer names overl