A client server model to facilitate creation of a medical image teaching library.

A simple and convenient system for indexing and archiving medical images used in teaching was developed. The approach was to combine a smart client-side graphical user interface that controlled image size, file format, and keyword structure, and communicated with the hospital information system via hypertext mark-up language, to populate the interface with user selectable pull-down menus. The result is a system that is easily extensible beyond the radiology images for which it was originally designed. Only minor modifications of the client interface are required to adapt the program to accept any file format or image type.

Development of an automated 12-8 bit conversion algorithm for displaying and archiving scanned radiographs.

The lossless capture of the entire range of optical densities encountered in a radiograph can require up to 4000 individual shades of gray. Most inexpensive computer displays, and most grayscale image formats contain only 256 gray shades. To use these systems, a look up table must be generated to display the scanned image, which maps the approximately 4000 gray shades in the digitized radiograph to the 256 gray shades on the monitor. However, if this conversion is inappropriately performed, much of the information contained in the original image could be lost. This study was undertaken to develop an histogram equalization algorithm that would automatically convert the 4000 gray shade digitized radiograph into a 256 gray shade image file, without perceptable loss of diagnostic information or image quality.