DEMONSTRATION

A Web-Based Image Bank to Consolidate Pathology Department Image Resources for Medical Education

Jannie Woo, Ph.D. and Robert Schelper, M.D., Ph.D. Department of Pathology, SUNY Upstate Medical University at Syracuse, New York, USA

ABSTRACT:

In keeping with most traditional long-standing medical institutions, the Department of Pathology at SUNY Upstate Medical University has accumulated large collections of glass slides and Kodachromes of macroscopic and microscopic images from various resources including tissues submitted for analysis, and from retired and current faculty members. Locating suitable images for teaching pathology residents and medical students is often times a challenge and frustration, since searching across multiple collections is time consuming and seldom productive; search criteria for images from varied resources are often incomplete, and search terms used are non-standardized.

To overcome this problem, we developed a Web-based image database to consolidate macroscopic and microscopic images from different resources within our own department. Images from glass slides and Kodachromes are digitized using a Nikon scanner and each image touched-up using PhotoShop. Each image is identified by a unique alphanumeric code and is categorized according to its image type, diagnosis, case number, organ type, and method of stain if microscopic. A brief annotation for each image is input with the help of willing faculty and residents. Copyright holders of images are also identified, and should any images be downloaded for PowerPoint lecturing use, appropriate image source acknowledgement is requested.

To encourage current faculty and residents to share their own collections, the image database is constructed to readily facilitate direct inputting of images and their attributes. For each image, the potential contributor simply enters the image description into a blank Web-based template with a unique alphanumeric code, which then becomes the image name when it is uploaded to the server via FTP. To provide security, the module is password protected. After signing on, the user may search using single or multiple search criteria to retrieve either a general or a specific list of images. Each image may be reviewed along with its categorized description, and may be downloaded as an 800x600 copy suitable for PowerPoint lecture display.

To date more than 6000 high quality retrievable images have been loaded into the module, which has shown to be invaluable for teaching faculty in need of specific images for teaching. Our current goal is to standardize image description terms, and to introduce metadata to improve the image retrieval process.

BENEFIT TO PARTICIPANTS ATTENDING SESSION:

Participants will learn from this demonstration:

· . How to set up a Web-based image database

· How to develop Web-based templates to allow direct input and edit of image information into the database

· How to retrieve information from a database for Web display

· The importance of standardized image descriptors for effective search and retrieval

Jannie Woo, Ph.D.

Department of Pathology

SUNY Upstate Medical University

Syracuse, NY 13210

USA

Phone: 315-464-6717

Fax: 315-464-7130

mailto:wooj@upstate.edu

Website: http://pathed.upstate.edu:8080/gripe/frame.htm

CO-AUTHORS:

Robert Schelper, M.D., Ph.D

Department of Pathology

SUNY Upstate Medical University

Syracuse, NY 13210

USA

Phone: 315-464-7167

Fax: 315-464-7130

mailto:schelper@upstate.edu

Website: http://pathed.upstate.edu:8080/gripe/frame.htm