OCT Segmentation

Optical Coherence Tomography (OCT) is a non-invasive technology used for imaging the layers of the retina. This imaging technology has revolutionized the early detection and also follow-up examination of retinal pathologies as as macular holes, epiretinal gliosis, macular edema and also optic disc pathologies. The introduction of high speed and spectral domain OCT devices allows a faster image acquisition and offers a better image resolution, an advantage for the daily clinical practice.

However, the continuous accession of image data available for diagnostic purpose demand sophisticated data processing strategies to help the Ophthalmologist to focus on potentially pathological structures while screening the huge amount of data. In addition, objective quantification of pathologies (thickness measurements, detection of changes in reflectivity of certain structures, loss of layers in degenereative pathologies) is an indispensable tool to diagnose disease progression and also regression and subsequently base treatment decisions on that.

The objectives of this project are twofold, first to implement, test, and validate new segmentation strategies to reliably identify individual structures of the retina. Second, localize potentially pathological areas within the retina that will need the attention of the investigator.

In general an automatic segmentation framework for OCT slice stacks will pave the way for a more systematic and objective data screening in clinical practice. Moreover it can help to prevent misinterpretation of images due to time constrains or missing experiences of the examiner.



The software may be used freely for research purposes. Of course, we ask you to cite our paper if you use the application in you publications:

P. A. Dufour et al.“Graph-Based Multi-Surface Segmentation of OCT Data Using Trained Hard and Soft Constraints.” IEEE Trans. Med. Img. 2012.

Download and Requirements:

  • Supports Heidelberg Spectralis OCT exports from Heidelber Eye Explorer
  • Windows 7, 64 bit install package, other architectures to follow

Downloads: Application Setup

We gonne make the application available for download soon. However, in a frame of a research collaboration you can start using the application immediately. Please get in contact with us. 

  • Windows 64bit: comming soon...
  • Windows 32bit: comming soon...
  • Mac OSX 64bit: comming soon...

Downlodas: Example File

  • 3x OCT volume of healthy subjects: zip file
  • 3x OCT volume of AMD patients: zip file

Downloads: Documentation

Manual(preliminary): OCTSegmentationUserManual.pdf (pdf, 4.6 MB)


Current Features

  • Automatic segmentation of 6 retinal cell layers in healthy eyes
  • Automatic segmentation of drusen in eyes of patients with AMD
  • Individual cell layer thickness measurements from ETDRS sectors and in specified reginos in B-scans
  • Manual Segmentation and Correction
  • Average cell layer reflectivity measurements
  • 3D visualization