METHODS OF GRAPH SEARCHING FOR BORDER DETECTION IN IMAGE SEQUENCES WITH APPLICATIONS TO CARDIAC MAGNETIC-RESONANCE-IMAGING

Automated border detection using graph searching principles has been s hown useful for many biomedical imaging applications, Unfortunately, i n an often unpredictable subset of images, automated border detection methods may fail. Most current edge detection methods fail to take int o account the added information available in a temporal or spatial seq uence of images that are commonly available in biomedical image applic ations, To utilize this information we extended our previously reporte d single frame graph searching method to include data from a sequence, Our method transforms the three-dimensional surface definition proble m in a sequence of images into a two-dimensional problem so that tradi tional graph searching algorithms may be used, Additionally, we develo ped a more efficient method of searching the three-dimensional data se t using heuristic search techniques which vastly improve execution tim e by relaxing the optimality criteria. We have applied both methods to detect myocardial borders in computer simulated images as well as in short-axis magnetic resonance images of the human heart. Preliminary r esults show that the new multiple image methods may be more robust in certain circumstances when compared to a single frame method and that the heuristic search techniques may reduce analysis times without comp romising robustness.