Histology Image Analysis for Cervical Carcinoma Detection and Grading
Brown Bag Lecture by Dr. Lei He | 4/26/2011 11AM-12PM | 7th Floor Conference Room, Bldg 38A
Abstract: This talk will introduce the application of image analysis techniques to the domain of histopathology, specifically, for the objective of cervical intraepithelia neoplasia (CIN) detection and classification. In the past decades, numerous computer assisted diagnosis (CAD) systems have been implemented to aid clinicians and oncologists in cancer diagnosis and research, which significantly reduced the labor and subjectivity of traditional manual intervention with histology images. A brief review of regular CAD system flowchart will be presented, which consists of procedures of image preprocessing, segmentation, feature extraction and dimension reduction, disease detection and grading, and postprocessing. In particular, emphasis is given to state-of-the-art image segmentation methods for feature extraction and disease classification. My recent algorithms specifically developed for histology image segmentation will be presented in the comprehensive comparison with other generalized models.
Bio: Dr. Lei He is a visiting faculty of the National Library of Medicine, in the National Institutes of Health. He works on histology image analysis for cervical cancer diagnosis. Before he joined the NIH in July 2009, he was a tenured associate professor of the Department of Computer Science and Information Technology at Armstrong Atlantic State University in Savannah, Georgia. His research interests include image processing and computer vision, machine learning and pattern recognition. He has published 1 book on image understanding, 2 book chapters of color image restoration and histology image analysis, about 50 refereed journal and conference papers, and secured 1 patent with the Hewlett Packard Laboratory. He has implemented a number of industry and medical projects with different R&D centers and hospitals.