Children's National Medical Center

Juan Jose Cerrolaza

Marius George Linguraru

Raymond W Sze

Laurie S Conklin

Craig Peters

Nabile M Safdar

Described is a method for segmenting an anatomical part, including identifying a landmark on a contour/surface of a 2D/3D model of the anatomical part in an ultrasound image, assigning a weight to the landmark, identifying different appearance patterns of a region around the landmark based on a previously stored training set; and applying a filter to the different appearance patterns of the region around the landmark in order to identify contours of the anatomical part.

Medical anatomy quantification: computer-aided diagnosis tool

Discussed herein is an image processing device and a corresponding method for assessing inflammatory bowel diseases. A plurality of dynamic MRI sequence of images of a bowel that include peristalsis information of the bowel and static MRI image(s) of the bowel that identify at areas of inflammation, stenosis, and increased mural thickness of the bowel are processed. A displacement of the bowel caused by respiration of the patient is conducted on the dynamic MRI sequences by performing a non-rigid registration. A peristaltic activity is determined by performing motion flow analysis on the dynamic sequences of images. A peristaltic activity of the static MRI image(s) based on a weighted sum of two closest dynamic MRI sequences of images is computed and the static image is combined with the closest dynamic MRI sequence of images by performing a multi-modal registration.

Growing community of inventors

Washington, DC, United States of America

Juan Jose Cerrolaza