THIS IS A SHANNON AWARD PROVIDING PARTIAL SUPPORT FOR THE RESEARCH PROJECTS THAT FALL SHORT OF THE ASSIGNED INSTITUTE'S FUNDING RANGE BUT ARE IN THE MARGIN OF EXCELLENCE. THE SHANNON AWARD IS INTENDED TO PROVIDE SUPPORT TO TEST THE FEASIBILITY OF THE APPROACH; DEVELOP FURTHER TESTS AND REFINE RESEARCH TECHNIQUES; PERFORM SECONDARY ANALYSIS OR AVAILABLE DATA SETS; OR CONDUCT DISCRETE PROJECTS THAT CAN DEMONSTRATE THE PI'S RESEARCH CAPABILITIES OR LEND ADDITIONAL WEIGHT TO AN ALREADY MERITORIOUS APPLICATION. THE ABSTRACT BELOW IS TAKEN FROM THE ORIGINAL DOCUMENT SUBMITTED BY THE PRINCIPAL INVESTIGATOR. DESCRIPTION: (adapted from the application abstract) The advent of large picture archiving and communication systems (PACS) will likely result in a conversion of clinical radiology to become nearly completely digital. Improved methods for access to a large collection of on-line image data can improve medical research and education. Although the techniques for fast text search are well established, similar tool development for rapidly searching through years of digitally archived images is still in its infancy. Development of a fast browsing technology for retrieving archived images over a network, both local and remote (international) accesses is proposed. In this project, three key technologies will be merged: smart query, hierarchical archive, and photo indexing using embedded zerotree wavelet transform (EZTWT) code. The first two technologies provides query constraint and automatic image migration management, and are supported in other PACS- related projects at UCLA. The third and newest component, EZTWT, is an encoding method designed for maximum network utility, providing the receiver highest image quality for a given transmission time. The implications for high-performance teleradiology under bandwidth limitations are significant. Diagnostic viewing can start before transmission of an image in full resolution and quality is complete. This encoding scheme will be modified to allow variable size iconic representation of the original, achieving minimum network delay and fast reconstruction, thus making browsing through a larger selection of images convenient and manageable. In addition, the modified EZTWT can achieve high efficiency for images with clearly separated anatomical and background regions. The space saving in very high resolution radiographs, such as mammograms, can be typically 7MB per 4Kx4K image (44%) without loss in anatomical information. This coding technique will be tested in conjunction with new pattern recognition techniques to achieve efficient mammogram storage and telecommunication.
