CEB offers summer and year-round employment opportunities for qualified student and postdoctoral applicants. Find information about summer student positions and postdoctoral positions at http://www.training.nih.gov. For year-round internships, send cv to Sheila Levy, email: slevy@mail.nih.gov.

Jing Cheng
I am a Ph.D. student with the Department of Electrical and Computer Engineering, University of Maryland, College Park. My research interests include image processing, computer vision and pattern recognition. I am working with Dr. Sameer Antani on the Advanced Medical Imaging Tools (AMIT) project. Specifically, I am developing software tools for CBIR of medical images over the Internet. The tools enable shape-based retrieval of digitized spine X-ray images from the Second National Health and Nutrition Examination Survey (NHANES II).The tools provide combined access to image and text databases. Currently, I am developing a medical image validation and pathology collection tool. The tool will enable radiologists and other medical experts in providing pathology information pertinent to our project and validate the segmented boundaries marked by (semi) automated methods operated by technicians. I also assist in the maintenance and management of CEB webpages.

Jonathan Long
Student, Thomas S. Wootton High School
I am working with Mukil under the direction of Dr. Sameer Antani on the next iteration of the CBIR system for spine X-ray image processing and analysis. Specifically, I am using Java and MATLAB to improve CBIR version 3’s image segmentation and indexing facility, and to add various useful capabilities, internally and externally, including XML export and import and a segmentation enhancement GUI. When completed, the system will allow users to perform computer assisted segmentation, indexing, and search of vertebra images in an integrated platform complete with remote database, various experimental techniques of segmentation and search, and full XML support.

Praveer Mansukhani
I am a PhD candidate in the Computer Science and Engineering Department at SUNY – University at Buffalo. In Summer 2005, I worked with Dr Song Mao on automated metadata extraction from historical documents. We used Support Vector Machines, followed by stochastic modeling to automatically classify text lines from scanned historical FDA court records. Metadata are extracted from labeled text lines using OCR and metadata specific keywords. The algorithms will be integrated into the SPER (System for Preservation for Electronic Resources) for easy access, indexing and retrieval of the documents over the long term.

Kiran K. Vasudevan
I am a graduate student working towards PhD degree in ECE at University of Maryland College Park. I am working on Interactive Publications project with Dr. Thoma, Dr. Antani, Glenn Ford and Michael Chung. The long term objective of our research is to develop a robust standard which would help in creating open source, platform independent tools for authoring, publishing and archiving Interactive Publications. The short-term objective is to demonstrate the creation and usage of Interactive Publications using existing standards and tools. I intend to develop my PhD thesis on one of the several hurdles in the development of Interactive Publications.

Xiaoqian Xu
I am a first-year graduate student at Brigham Young University majoring in Electrical Engineering. I have a summer position this year at CEB of the National Library of Medicine. And I have been working on Partial Shape Matching (PSM) algorithm for vertebra shapes with Dr. Sameer Antani. We also develop GUI for using PSM algorithm based on the old CBIR system. So by adding PSM option, the new interface allows the user to query by complete shape as well as partial shape. I really enjoyed this summer. It has been such a great working experience and a fond memory as well.

Jian Yao
Ph.D candidate, Electrical and Computer Engineering, State University of New York at Binghamton
My research areas include computer vision, pattern recognition, machine learning, and data mining. I am working with Dr. Sameer Antani for automatic medical image annotation project. Unlike the existing annotation methods, which are achieved through direct classification, our method is realized by indirect classification. This indirect annotation method is expected to be included into image retrieval algorithms, such as GIFT.