Artificial neural networks and spatial temporal contour linking for automated endocardial contour detection on echocardiograms: A novel approach to determine left ventricular contractile function

This study investigated the use of artificial neural networks (ANN) for ima ge segmentation and spatial temporal contour linking for the detection of e ndocardial contours on echocardiographic images. Using a backpropagation ne twork, the system was trained with 279 sample regions obtained from eight t raining images to segment images:es into either tissue or blood pool region . The ANN system was then applied to parasternal short axis images of 38 pa tients. Spatial temporal contour linking was performed on the segmented ima ges to extract endocardial boarders, Left ventricular areas (end-systolic a nd end-diastolic) determined with the automated system were calculated and compared to results obtained by manual contour tracing performed by two ind ependent investigators, In addition, ejection fractions (EF) were derived u sing the area-length method and compared with radionuclide ventriculography , Image quality was classified as good in 12 (32%), moderate in 13 (34%) an d poor in 13 (34%) patients, The ANN system provided estimates of end-diast olic and end-systolic areas in 36 (89%) of echocardiograms, which correlate d well with those obtained by manual tracing (R = 0.99, SEE = 1.44), A good agreement was also found for the comparison of EF between the ANN system a nd Tc-radionuclide ventriculography (RNV, R = 0.93, SEE = 6.36), The ANN sy stem also performed well in the subset of patients with poor image quality. Endocardial contour detection using artificial neural networks and spatial temporal contour linking: allows accurate calculations of ventricular area s from transthoracic echocardiograms and performs well even In images with poor quality, This system could greatly enhance the feasibility, accuracy a nd reproducibility of calculating cardiac areas to derive left ventricular volumes and ejection fractions, (C) 1999 World Federation for Ultrasound in Medicine gr Biology.