We specialise in a number of medical applications of computer vision. We have a strong track record in retinal image processing, having developed techniques to support automated diagnosis of retinal diseases such as diabetic retinopathy, sponsored primarily by Diabetes UK. We are currently developing new model-based methods for vascular segmentation and measurement, based on active contour models, and for lesion segmentation based on level-sets. The accuracy of these new methods will lead in time to better diagnostic capability. We are also active in the application of surveillance and tracking techniques to medical and social domains, including: the use of pose and movement analysis to identify episodes of sleep apnoea, and the use of tracking and state modelling techniques to provide autonomous care monitoring to vulnerable individuals, including the elderly, at home.
Our work in diabetic retinopathy includes the REVIEW database http://reviewdb.lincoln.ac.uk a public-domain database of retinal vessel segments, including reference standard retinal vessels marked by clinicians.
Investigators
- Beth Massey
- Ching-Wei Wang
- Bashir Al-Diri
- Andrew Hunter
External Cooperation
- United Lincolnshire United NHS Trust
- Sunderland Eye Infirmary
Representative Publications
Al-Diri, B., Hunter, A., Steel, D. and Habib, M. (2008): Joining Retinal Vessel Segments, 8th IEEE International Conference on BioInformatics and BioEngineering (BIBE 2008), Athens, Greece, October 8-10.
Wang, C.-W. and Hunter, A. (2008): A Robust Pose Matching Algorithm for Covered Body Analysis for Sleep Apnea, Proceedings of the 8th IEEE International Conference on Bioinformatics and Bioengineering (IEEE BIBE 2008), Athens, Greece, 8-10 October.
Lowell, J., Hunter, A., Steel, D., Basu, A., Ryder, R., Fletcher, E. and Kennedy, R.L. (2004): Optic nerve head segmentation, IEEE Transactions on Medical Imaging, 23, 2, pp. 256-64.
Lowell, J., Hunter, A., Steel, D., Basu, A., Ryder, R. and Kennedy, R.L. Measurement of Retinal Vessel Widths from Fundus Images Based on 2-D Modeling, IEEE Transactions on Medical Imaging, 23 10.
Search Publications using ePrints: http://eprints.lincoln.ac.uk