Electronic Stool Subtraction Using Quadratic Regression and Intelligent Morphology

Reference #:

2007-008

Inventors/Contributors

Armando Manduca Ph.D., Michael J. Carston, C. Daniel Johnson M.D.

Description

CT colonography (CTC) is being extensively studied for its potential value in colon examinations. Like traditional colonoscopy, CTC requires extensive preparation to cleanse the colon of fecal material. The inconvenience and discomfort associated with this preparation is an important obstacle to compliance with currently recommended colorectal screening guidelines. An alternative is to tag the stool with an opacifying agent so that it is brighter than soft tissue and thus recognized automatically. Pixels identified as stool can then be electronically subtracted away by being reset to air values, permitting rapid interpretation of the CTC dataset and 3D endoluminal images. This invention consists of a novel technique for electronically removing fecal material from CT colonography data. The algorithm identifies transitions between material types (stool, air, or tissue) by examining the relationship between the intensities and gradients of a given voxel and its neighbors. This relationship is approximately parabolic, with different scales and locations for each transition type. The appropriate transition type is classified using a quadratic least squares regression. Intelligent morphological image processing techniques are used to improve the classifications.

Patent Status

Pending