Non-intrusive retinal imaging system

Developed by Prof. Jane You Jia of the Department of Computing, the novel Retinal imaging System performs computer-aided, non-intrusive diabetic retinopathy (DR) screening and monitoring and affords privacy protection.

The special features and advantages of the system include:
• A special optical design for a new fundus camera model that allows lighting normalization and position alignment;
• A hierarchical approach to retinal image analysis via multiple feature extraction and coarse-to-fine lesion detection; and
• A wavelet-based scheme for lossless data embedding to facilitate information security and privacy protection.

The system demonstrates effective diabetic care, and has excellent potential for widespread applications, as described in the following.

• The new effective retinal image analysis software in conjunction with the high-performance fundus camera has widespread clinical applications, not only by ophthalmologists in eye care, but also by other medical specialists in the provision of general health services, including telemedicine.
• The new algorithms for lossless data embedding will make a significant contribution to information technology applications, ranging from multimedia to information security.

This innovation won a Gold Award with the Congratulations of the Jury at the 39th international Exhibition of Inventions in Geneva, Switzerland. It also took Second Place in the SPIE Medical Imaging 2009 International Competition – Retinopathy Online Challenge.

This article was first appeared on the PolyU Milestone, June 2011 Edition