Ophthalmic imaging

Ophthalmic imaging is a rapidly growing field of research. Not only new imaging modalities found their way into clinical examination routines but also the need to support physicians by visualization and image processing techniques has increased. The goal of the Ophthalmic Imaging Group is to develop algorithms in the field of pattern recognition as well as image processing to support physicians. Our methods provide objective measurements to aid diagnosis and enable fully automatic screening of eye diseases like glaucoma or diabetic retionopathy.

Current Research Topics:

Fundus imaging is the most commonly used modality by ophthalmologists for non-invasive diagnosis of eye diseases. Fundus cameras provide 2-D photographs of the human eye background. Anatomical structures such as blood vessels or the optic nerve head are analyzed for computer assisted diagnoses.

Our research covers the pattern recognition pipeline from image preprocessing to classification:

  • Image restoration for low-cost cameras with poor image quality
  • Segmentation and analysis of anatomical structures
  • Classification techniques for computer assisted diagnoses

Optical Coherence Tomography (OCT) is a non-invasive imaging technique that is able to capture 2-D and 3-D data sets of transparent tissue such as the retina with micrometer resolution. Recently, OCT has found widespread application in clinical practice and has become a standard technique for assessing eye diseases. In order to facilitate computer aided diagnosis it is necessary to assure sufficient data quality and to automatically extract quantitative measures.

Our research covers the following topics:

  • OCT motion correction and signal enhancement (see figure)
  • Automatic segmentation of OCT image data
  • Registration of OCT volumes and fundus images

En face fundus projection (left) and OCT volume (right) before motion correction (top row) and the result of OCT motion correction (bottom row).