ROBUST SIMULTANEOUS DETECTION OF CORONARY BORDERS IN COMPLEX IMAGES

Visual estimation of coronary obstruction severity from angiograms suf fers from poor inter- and intraobserver reproducibility and is often i naccurate, In spite of the widely recognized limitations of visual ana lysis, automated methods have not found widespread clinical use, in pa rt because they too frequently fail to accurately identify vessel bord ers, We have developed a robust method for simultaneous detection of l eft and right coronary borders that is suitable for analysis of comple x images with poor contrast, nearby or overlapping structures, or bran ching vessels, The reliability of the simultaneous border detection me thod and that of our previously reported conventional border detection method were tested in 130 complex images, selected because convention al automated border detection might be expected to fail, Conventional analysis failed to yield acceptable borders in 65/130 or 50% of images , Simultaneous border detection was much more robust (p < .001) and fa iled in only 15/130 or 12% of complex images, Simultaneous border dete ction identified stenosis diameters that correlated significantly bett er with observer-derived stenosis diameters than did diameters obtaine d with conventional border detection (p < 0.001), Simultaneous detecti on of left and right coronary borders is highly robust and has substan tial promise for enhancing the utility of quantitative coronary angiog raphy in the clinical setting,