Endoscopic Multimedia Information System

 

Personnel

Dr. Wallapak Tavanapong, Associate Professor of Computer Science, Principal Investigator

Dr. Johnny Wong , Professor and Associate Chair Dept. of Computer Science , Co-Principal Investigator

 

Summary

Advances in video technology are being incorporated into today's healthcare practice. For example, various types of endoscopes are used for colonoscopy, upper gastrointestinal endoscopy, enteroscopy, bronchoscopy, and cystoscopy. In addition, a rapidly expanding number of formerly open surgical procedures now are being converted to endoscopic procedures including resection of gallbladders, retrieval of donor kidneys, resection of tumors of colon and pancreas, correction of hiatal hernias, coronary artery bypass grafting and minimal invasive neurosurgeries. Despite a large body of knowledge in medical image analysis, endoscopy videos are not systematically captured for real-time or post-procedure reviews and analyses. They are recorded occasionally to magnetic video-tapes (i.e., VHS). No hardware and software tools have been developed to capture, analyze, and provide user-friendly and efficient access to the medical, scientific, or educational content on such videos. This project aims to develop an Endoscopic Multimedia Information System (EMIS) to capture high quality endoscopy videos, analyze the captured videos, and provide efficient access to the content of these videos. Images of endoscopy videos significantly differ from medical still images as studied in the literature of medical image processing. The project will develop a new capturing system for endoscopy videos designed for patient's privacy and is non-disruptive to endoscopic procedures and non-restrictive to a particular endoscope vendor and new algorithms for automatic classification of informative and non-informative frames and new algorithms for automatic content analysis for protruding lesions such aspolyps. Many protruding lesions are clustered together. A new region segmentation technique that can identify isolated lesions will be developed first. To handle clustered lesions, algorithms using a region pattern graph that captures important characteristics of relevant regions will be developed.. The proposed system will directly benefit endoscopic research, education, and training.

 

Funding

This research project supported in part by a National Science Foundation Grant SEI 0513809.