医学院(生物医学工程系)学术报告：

简介：

Colonoscopy is a medical procedure where a long, flexible, tubular instrument called a colonoscope is used to view the entire inner lining of the colon to identify possible polyps and colorectal cancer. During conolosopy, as the doctor uses one hand to push (pull) the colonoscope into (out of) the colon, he must use the other hand to rotate the two knobs on the colonoscope to adjust the tip orientation to make the scope point to the cavity of the colon (lumen) through his evaluation of the colon video. This manual procedure is normally onerous with about 0.1% perforation rate and 0.052% death rate. In order to make this task easy and minimize the death risk, in this work, an image guidance control system for colonoscopy was investigated. Through the automatic processing of colon images, the lumen positions of the colon were obtained. The positions were then converted to the relative adjustment angles of the bending section (tip) of a colonoscope. With a simulated dynamic model of the bending section used in the current colonoscopy, the optimal motion control was achieved. By feeding the controller with the obtained adjustment angles as the error information relative to the center of the colon images, the tip of the conoloscope can be controlled to point to the lumen during a colonoscopy to realize the lumen tracking motion. Furthermore, in order to prevent perforation, a shape-sensor system was investigated. Both lab and field tests were carried out to verify the functionality of the system. This technique can lead to better accuracy on colon lining inspection in diagnostics and operations, such as biopsy. By automating tip orientation, it is expected to reduce the training time for a new doctor, improve the efficiency of the operations, and reduce surgical accidents.

---------

Debao's Bio:

Dr. Debao Zhou is now with the Georgia Tech Research Institute as a Research Engineer. He received his Ph.D. degree in Mechanical Engineering from Nanyang Technological University, Singapore. His thesis topic is on the force and motion control of a walking robot for terrain adaptation. He earned the M. Eng and B. Eng degrees from Tsinghua University, P. R. China, both in Mechanical Engineering. His research interests include system dynamics, bio-material modeling, computer vision, bio-medical robotics, non-linear control system. He is a member of the Institute of Electrical and Electronics Engineers (IEEE), the American Society of Mechanical Engineers (ASME), and the International Association of Science and Technology for Development (IASTED).