Mean Shift Framework for Robust Analysis of Lung Nodules in CT Scans

日時:
2008年1月11日 15:00-16:15

場所:
吉田キャンパス工学部３号館北棟１階　Ｎ１講義室

講師:
Kazunori Okada氏

講師所属:
Ph.D., Computer Science Department, San Francisco State University

講師略歴:
Dr. Okada has broad research interests in the areas of intelligent computing, such as computer vision, pattern recognition, machine learning, artificial intelligence and data mining. He has been active in the research fields of medical image analysis, statistical data analysis, cognitive vision and face recognition. His earlier work on face recognition has produced a winning system in the well-known FERET competition, setting the industry-standard. His recent work on lung tumor segmentation and detection in chest CT scans has resulted in a number of US, German, Chinese and Japanese pending patents. He has received the Ph.D. and M.S. degrees in computer science from University of Southern California, and the M.Phil. degree in human informatics and the B.Eng. degree in mechanical engineering both from Nagoya University in Japan. He is currently an assistant professor of computer science at San Francisco State University and leads the laboratory for biomedical data analysis. Prior to his academic appointment, he was a research scientist at Siemens Corporate Research in Princeton, NJ. He is a member of IEEE, ACM, SPIE and MICCAI.

講演概要:
In past decades, advancements of medical imaging devices have drastically improved their qualities by rapidly increasing image resolutions and data volumes, aiming for better diagnostic accuracy. However, this also caused an exponential increase in the amount of data that radiologists must study, therefore imposing ever-increasing labor burdens and causing potential misdiagnoses due to fatigues. Medical Image Computing has been a potential answer to such critical demands. In this seminar, I will present a body of work, conducted at Siemens Corporate Research, Princeton, USA, focusing on making such medical image computing more robust, efficient, and interactive, in order to meet the above clinical and technical demands. The well-known mean shift technique and corresponding robust kernel-based statistical modeling have been extended toward clinical application for detecting, segmenting and classifying the lung cancers in volumetric CT scan data. This seminar will outline motivations and advantages for such robust and interactive data analysis techniques. As a concrete example, I will present my work on robust Gaussian fitting and scale selection for segmenting clinically-significant and technically-challenging ground-glass-nodule (GGN) cases.

備考:
4th International Seminar on Primordial Knowledge Modelを兼ねています．
http://blog.ii.ist.i.kyoto-u.ac.jp/?p=197

連絡先: 牧 淳人
所属: 情報学研究科知能情報学専攻
メールアドレス: maki@(at).kyoto-u.ac.jp