We present here a new algorithm for segmentation of nuclear medicine images to detect the left-ventricle (LV) boundary. In this article, other image segmentation techniques, such as edge detection and region growing, are also compared and evaluated. In the edge detection approach, we explored the relationship between the LV boundary characteristics in nuclear medicine images and their radial orientations: we observed that no single brightness function (eg, maximum of first or second derivative) is sufficient to identify the boundary in every direction. In the region growing approach, several criteria, including intensity change, gradient magnitude change, gradient direction change, and running mean differences, were tested. We found that none of these criteria alone was sufficient to successfully detect the LV boundary. Then we proposed a simple but successful region growing method~Contour-Modified Region Growing (CMRG). CMRG is an easy-to-use, robust, and rapid image segmentation procedure. Based on our experiments, this method seems to perform quite well in comparison to other automated methods that we have tested because of its ability to handle the problems of both Iow signal-to-noise ratios (SNR) as well as Iow image contrast without any assumptions about the shape of the left ventricle. bounda¡ 1-3 Correct and reproducible measurements of LV function, in terms of parameters such as the global and regional ejection fraction, require an accurate and reproducible algorithm to delineate the LV. 4 6 Delineation can be accomplished manually or automatically using an edge detection algorithm. Manual algorithms usually suffer from low reproducibility, whereas automatic algo¡ suffer from low precision because of nonuniform background, low signal-to-noise ratios (SNR) and the complicated structure of the heart. 6 In most clinical applications, manual tracking or semiautomated boundary description is conducted in each 2D nuclear medicine image by a trained clinician to extract the LV contour. Although the reliability of the edited contour can be enhanced by introducing more skilled operators anda priori knowledge (such as location, shape, and intensity), manual tracking has two major drawbacks. First, the obtained LV boundaries are biased (the tracked boundaries vary from individual to individual). Second, manual tracking is extremely time-consuming. 3To resolve the two disadvantages of manual tracking, various automated or semiautomated techniques of varying complexity and precision previously have been proposed. 2,s-6 A comparison of three major commercially available semiautomatic methods was provided by Bingham et al. 7 A comprehensive review of the computer methods for quantitative analysis of LV function from equilibrium gated blood pool scintigrams was given by Reiber et al. 8 The most commonly used edge detection algorithm is based on the assumption that the LV border coincides with the zero crossing of the second-order derivative in a radial search with its origin in the LV, according to Reiber et al. 8 To...