The overall objective of the proposed research is to develop a clinically useful, computer-based methodology to assist in decision-making tasks in medical imaging. The methodology consists of a novel approach combining well-established mathematical methods and artificial intelligence techniques for simultaneously representing medical knowledge and handling visual information.
The specific aims of the proposed research are to develop, implement, test, and validate a knowledge-based system for interpreting three-dimensional thallium-201 myocardial distributions obtained by tomographic imaging. These interpretations will assist clinicians in detecting and localizing coronary artery disease. The research will be carried out in a collaboration between Emory University and the Georgia Institute of Technology, thus providing a unique, interdisciplinary, and interinstitutional approach that directs complementary clinical and scientific resources toward this important medical problem. The five-year research program consists of four projects, dealing with (1) knowledge representation, (2) image understanding, (3) clinical acceptance and validation, and (4) system extension and refinement. Preliminary studies indicate that the methodology can provide both accurate cardiac image interpretations and graphical, computer-generated representations of arterial structure. The approach can serve as a clinically useful model with applications to other imaging modalities, and can also be used as a tool for tutoring and training clinicians. The knowledge-based system is designed to be sufficiently extensible and flexible to accommodate the far-term objectives of eventually providing decision-making support to clinicians in tasks associated with diagnosis, therapy, and prognosis.
| Peifer, J W; Ezquerra, N F; Cooke, C D et al. (1990) Visualization of multimodality cardiac imagery. IEEE Trans Biomed Eng 37:744-56 |
| Klein, J L; Garcia, E V; DePuey, E G et al. (1990) Reversibility bull's-eye: a new polar bull's-eye map to quantify reversibility of stress-induced SPECT thallium-201 myocardial perfusion defects. J Nucl Med 31:1240-6 |
| DePuey, E G; Garcia, E V; Ezquerra, N F (1989) Three-dimensional techniques and artificial intelligence in thallium-201 cardiac imaging. AJR Am J Roentgenol 152:1161-8 |
