patents

2003	Segmentation of ventriculograms using boosted decision trees Inventors: John Alan McDonald; Florence H. Sheehan An automated method for determining the location of the left ventricle at user selected end diastole (ED) and end systole (ES) frames in a contrast enhanced left ventriculogram. Locations of a small number of anatomic landmarks are specified in the ED and ES frames. A set of feature images is computed from the raw ventriculogram gray level images and the anatomic landmarks. Variations in image intensity caused by the imaging device used to produce the images are eliminated by de flickering the image frames of interest. Boosted decision tree classifiers, trained on manually segmented ventriculograms, are used to determine the pixels that are inside the ventricle in the ED and ES frames. Border pixels are then determined by applying dilation and erosion to the classifier output. Smooth curves are fit to the border pixels. Display of the resulting contours of each image frame enables a physician to more readily diagnose physiological defects of the heart. Filed July 2003
2002	Automated delineation of heart borders and surfaces from images A method for fitting a surface to some portion of a patient's heart. In the method, ultrasound imaging is carried out over at least one cardiac cycle, providing a plurality of images in different image planes made with a transducer at known positions and orientations. An operator selects points on some of the images that correspond to the surface of interest, and a surface is automatically fit to the points in three dimensions (3D), using prior knowledge about heart anatomy to constrain the fitted shape to a reasonable result. The operator reviews the fitted surface, in 3D or alternatively, as intersected with the images. If the fit is acceptable, the process is done. Otherwise, the image processing is repetitively carried out, guided by the fitted surface, to produce additional data points, until an acceptable fit is obtained. The resulting output surface can be used in determining cardiac parameters. Inventors: Richard K. Johnson; John Alan McDonald; Florence H. Sheehan. Filed August 2002
1999	Reconstruction of three-dimensional objects using labeled piecewise smooth subdivision surfaces A method is disclosed for reconstructing a surface of an object using three-dimensional imaging data. While the method can be applied to reconstruct the surface of almost any object represented by three-dimensional data obtained by some imaging modality, an example of the technique is disclosed that reconstructs the surface of a left ventricle. In the disclosed example, digital imaging data for a patient's heart are traced producing a data set of points that define the border and specific anatomic features of the left ventricle. An abstract model of a generalized left ventricle is generated that fits a wide range of sizes and shapes of this portion of the heart. The abstract model includes an abstract control mesh in which the anatomic features are labeled and sharp (edge) characteristics are identified. Coordinates are assigned to the abstract control mesh, producing an initial embedded mesh, which is then subdivided twice to increase its smoothness. The embedded subdivided mesh is rigidly aligned with the data set points of the patient's left ventricle, and in particular, with the anatomic features. Finally, the aligned subdivided mesh is optimally fit to the data set points and anatomic features, yielding the reconstructed surface of the organ. The reconstruction of a surface using medical imaging data is a particularly good exemplary application of the invention, because it shows that the method can be employed for reconstructing the surface of any three-dimensional object that is defined by relatively sparse and noisy data points. Inventors: John Alan McDonald; Florence H. Sheehan; Edward L. Bolson; Malcolm E. Legget. US Patent 5,889,524
home	2003-11-17