awesome-oct

A curated list of open-source projects for Optical Coherence Tomography (OCT), covering:

• Data parsing
• Annotation
• Deep learning segmentation
• Format conversion
• Visualization

Goal: help researchers and engineers quickly find reusable OCT tooling.

---

Table of Contents

• Project List
• How to Choose
• Recommended Workflow
• Contributing

---

Project List

Category	Project	Description
Segmentation	OCTDL	Deep learning framework for OCT segmentation (training + inference)
Segmentation	keras-UNET-OCT	Lightweight UNet implementation in Keras
Segmentation	OCT-Retinal-Layer-Segmenter	Retinal layer segmentation (ILM, RPE, etc.)
Segmentation	eyeseg	Advanced retinal layer and fluid segmentation
Annotation	OCTAnnotate	OCT image annotation tool
Visualization	eyelab	Interactive visualization and labeling UI
Data Parsing	eyepy	Python library for Heidelberg .e2e data
Data Parsing	heyexReader	Low-level Heyex OCT reader
Data Parsing	LibE2E	C++ E2E parser for production systems
Conversion	OCT-Converter	Convert OCT data to images / NumPy / DICOM
Conversion	OCTA_DICOM2IMARIS	Convert OCTA DICOM to Imaris format

---

How to Choose

1) Your goal is to read raw data

Start with: eyepy, heyexReader, LibE2E.

2) Your goal is segmentation modeling

Start with: OCTDL, keras-UNET-OCT, eyeseg.

3) Your goal is annotation and quality review

Start with: OCTAnnotate, eyelab.

4) Your goal is format interoperability for downstream analysis

Start with: OCT-Converter, OCTA_DICOM2IMARIS.

---

Recommended Workflow

1. Parse source data with eyepy / heyexReader / LibE2E.
2. Convert into common formats with OCT-Converter (images or NumPy arrays).
3. Annotate training sets with OCTAnnotate or eyelab.
4. Train and run inference with OCTDL / keras-UNET-OCT / eyeseg.
5. Visualize and review outputs with eyelab.

---

Contributing

PRs are welcome. Please include:

• Project link (GitHub)
• Category (Segmentation / Annotation / Data Parsing / Conversion / Visualization)
• One-sentence summary (feature, use case, or highlight)

If you want to add a new category, include a short rationale in your PR.