OCTolyzer: Fully automatic toolkit for segmentation and feature extracting in optical coherence tomography and scanning laser ophthalmoscopy data

TL;DR

OCTolyzer addresses the need for transparent, reproducible, and scalable analysis of OCT/SLO ocular images by delivering an open-source toolkit with two analysis suites (OCT and SLO) for automatic segmentation and feature extraction. It leverages established segmentation models for choroid and vessels (Choroidalyzer, DeepGPET) and localiser SLO analysis, enabling thickness, area, CVI, and vascular metrics across macular and peripapillary regions. The authors demonstrate high reproducibility across multiple cohorts and data types, with low measurement noise relative to population variability and efficient runtimes on standard hardware. The work highlights the importance of open, validated pipelines for oculomics and provides a foundation for standardized ocular measurements in research and clinical translation.

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.

Figures (15)

• Figure 1: OCTolyzer’s pipeline. (A) Input: OCT (and optional SLO) image data with necessary metadata. (B -- D) OCT analysis suite for single/radial macular B-scans (B), macular volume scans (C), or peripapillary B-scans (D). (E) SLO analysis suite (SLOctolyzer burke2024sloctolyzer) for macula- or disc-centred localiser images.
• Figure 2: Diagram of single macular OCT B-scan measurements. (A) Localiser SLO showing the OCT acquisition line (green). (B) Horizontal-line OCT B-scan overlaid on the localiser. (C) OCT B-scan with retinal and choroid segmentations labelled (retinal layer definitions in Supplementary \ref{['supptab:layers_computed']}). (D) Thickness measurements drawn per A-scan for retina (top) and perpendicular to upper boundary for choroid (bottom). The solid black line indicates subfoveal thickness.
• Figure 3: Measurement process for a macular OCT volume. (A) 3D visualisation of an OCT volume scan. (B) Sequential B-scans with retinal and choroid segmentations. (C) Thickness maps for the inner, outer, and whole retina (top), and choroid thickness, vessel density, and CVI (bottom), with average ETDRS grid measurements overlaid. Inner retinal layer thickness maps can also be computed.
• Figure 4: Measurement process for an OCT peripapillary B-scan. (A) Circular OCT B-scan overlaid on the localiser SLO. (B) B-scan with retinal and choroid segmentations. (C) Detection of the temporal sub-field centre. (D) Division into 6 peripapillary sub-fields with retinal measurements overlaid. (E) Peripapillary retinal thickness profile with sub-field thresholds superimposed.
• Figure 5: Feature extraction for the localiser SLO image. (A) Disc-centred localiser SLO with segmentations and regions of interest overlaid. (B) Region of interest masks: whole image (green), zone C (blue), and zone B (red), with vessel features indicated by colour-coded arrows extending to the regions which they are measured in. (C) Flowchart of vessel metrics by segmentation map. Figure reproduced and edited with permission from Burke et al. burke2024sloctolyzer.