Heart disease is a vital health care problem, affecting millions of Americans each year. Clinical decisions regarding the diagnosis of heart disease rely on determining the amount of diseased myocardium,with myocardial perfusion imaging representing the most widespread clinical procedure for assessing myocardial infarction and/or ischemia. However, interpreting this image information, and integrating it with other clinical data, remains a difficult and ill-defined problem. With this in mind, this competing renewal application proposes to continue a research program with the overall objective of developing a clinically useful, computer-based methodology to aid in the diagnosis of heart disease. The methodology consists of a novel framework combining well established mathematical methods, visualization techniques, and artificial intelligence approaches for representing medical knowledge and integrating visual, numeric, textual, and temporal information. The principal hypothesis underlying the research is that medical decision-making tasks involving multidimensional information can be facilitated through the integration of both basic and applied concepts of medical informatics. During this next funding period, we propose to continue focussing our efforts on the development, implementation, and validation of this methodology through these specific aims: (1) automatic determination of the orientation of the left ventricular myocardium; (2) extension and enhancement of a knowledge base for interpreting myocardial perfusion imagery and other relevant information; (3) prediction of resting perfusion from resting thickening distributions through connectionist methods; (4) integration of connectionist and symbolic methods; (5) implementation and automation of the methodology into a fully integrated system; and (6) clinical testing and evaluation of this system. The extensive technical progress thus far achieved in these aims during the initial funding period is strong evidence of the merits of this highly interdisciplinary and interinstitutional research program.
