Project Proposal - Virtual Colonography

Project Proposal - Virtual Colonography - Virtual...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Virtual Colonography Baaba Andam, Siddhartha Ghosh, Albert Ho Abstract Virtual colonography is a non-invasive method of obtaining computerized images of the colon for analysis. This can be used to detect polyps in the colon, which if present, may lead to cancer. Our project will focus on implementing an algorithm to accurately detect polyps from CT scans of the colon. We will utilize a computer-aided diagnosis algorithm which involves computing the three dimensional geometric features and segments and clusters polyp candidates. We will compare our results to the ground truth and evaluate the performance of our algorithm by drawing an FROC curve. 1. Introduction Traditional colonoscopy involves inserting a colonoscope into a patient in order to get sample cells for analysis and detection of anomalies. Virtual colonography offers a non-invasive alternative to this. Computed tomography (CT) scans of the colon are taken and algorithms are used to detect polyps. Alternatively, the patient could swallow a pill with a camera which takes images of the colon for analysis. The main problem in polyp detection is that certain folds on the colon wall resemble polyps, leading to false positive detections in the implementation of polyp detection algorithms. On the other hand, some polyps do not protrude prominently from the colon wall, leading to false negative detections. We propose to implement the 3-D computer-aided diagnosis algorithm used by H. Yoshida and J. Näppi [3]. The paper by Yoshida and Näppi on the scheme for detection of colonic polyps will serve as our foundation study. Yoshida and Näppi have described a method that uses axial CT images, segments the entire colon and then detects polyp candidates, all automatically without human intervention. The paper also includes a section on methods of false positive reduction through the computation of 3D volumetric features that characterize the internal structures of the polyp candidates. However, the portions leading to the detection of candidates will be the relevant techniques for us, initially. The scheme developed by Yoshida and Näppi is unique because it extracts a thick region that encompasses the entire colonic wall,
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 09/24/2007 for the course ECE 5780 taught by Professor Reeves during the Spring '07 term at Cornell.

Page1 / 3

Project Proposal - Virtual Colonography - Virtual...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online