Coronary artery disease remains a major public health problem worldwide. It causes approximately 1 of every 6 deaths in the United States. Imaging of myocardial perfusion (delivery of blood to the heart muscle) by myocardial perfusion single photon emission tomography (MPS) allows physicians to detect disease before a heart attack, and predict risk in millions of patients annually. This is currently limited by the need fo visual interpretation, which is highly variable and depends on the physician's experience. The long-term objective of this program is to improve the interpretation of this widely used heart imaging technique-achieving higher accuracy for disease detection than it is possible by the best attainable visual analysis. This proposal builds on our prior work in conventional myocardial MPS, and focuses on fast, low-radiation MPS imaging (fast-MPS) obtained by new high-efficiency scanners. Specifically, we aim to: 1) develop new image processing algorithms for a fully automated analysis of fast-MPS. The algorithms will include better heart muscle detection by training with correlated anatomical data and a novel approach for mapping the probability of abnormal perfusion for each location of the heart muscle; 2) enhance the diagnosis of heart disease from fast-MPS by machine- learning algorithms that integrate clinical data, stress test parameters, and quantitative image features; 3) demonstrate the clinical utility of the new algorithms applied to automatic canceling of the rest portion of the MPS scan, when not needed. The new system will be more accurate than the clinical expert analysis in the detection of obstructive coronary disease. By immediately indicating whether a stress scan is normal, the system will allow for the automatic cancellation of the rest imaging portion when it is not needed (estimated in over 60% of all MPS studies). Our research will demonstrate that the computer decision regarding rest-scan cancellation is safe for the patient, both from a diagnostic and prognostic standpoint. This will lead to a wide adoption of low-dose stress-only imaging for MPS studies, which would reduce the amount of radiation that patients are exposed to, and allow for significant healthcare savings. It will additionally lead to a paradigm shift in the practice of nuclear cardiology, which will ultimately result in better selection of patients who need intervention, and reduce the number of deaths due coronary artery disease.
Imaging of myocardial perfusion (heart muscle blood flow) at rest and stress allows physicians to detect disease and predict risk in millions of patients in the US each year, but it is currently limited by the need of visual interpretation, which is dependent on doctor's experience. The investigators propose to develop and validate an automated, highly-accurate and objective computer system which will outperform even experienced physicians in interpreting these images using latest generation scanners and novel machine learning computer tools. The computer will be able to better select patients needing treatment and automatically indicate the normal stress-scan immediately, and more accurately than physicians allowing automatic cancellation of the rest imaging, when not needed; this will result in large healthcare savings and reduced radiation to the patients.
Showing the most recent 10 out of 67 publications
