City Research Online

Abstract

Medical conditions affecting the colon are problematic to diagnose due to the difficulty in examining this particular internal organ. To date, the most widely used approach is to perform a colonoscopy; a procedure in which a small camera is inserted into the colon to examine its surface. This procedure is unpleasant and potentially dangerous for the patient, and is expensive and time consuming for the hospital. As a result, patients at risk of developing the conditions are not always screened as often as would be desirable.
Over the last few years a new approach known as virtual colonoscopy has been gaining popularity. The method uses information from a CT scan to reconstruct a 3D model of the colon which can then be examined without the patient needing to undergo a colonoscopy. This approach is now commonly used when screening for polyps (an indication of colon cancer) but can not be so easily used on conditions such as Inflammatory Bowel Disease (IBD) where information beyond the shape of the surface is required.

This thesis forms part of a larger project which aims to diagnose conditions such as IBD by using image processing algorithms on CT data and presenting the results to the user in an easy to interpret way. Specifically we are concerned with this visualisation stage of the system and so have developed a new visualisation approach which we call Volumetric CPR. This can be used to supplement the more traditional virtual flythrough visualisation and is applicable to IBD detection as well as screening for polyps.
Our technique builds on the concept of Curved Planar Reformation (CPR), which has proved to be a practical and widely used tool for the visualisation of curved tubular structures within the human body. It has been useful in medical procedures involving the examination of blood vessels and the spine. However, it is more difficult to use it for structures such as the colon because abnormalities are smaller relative to the size of the structure and may not have such distinct density and shape characteristics.

Our new approach improves on this situation by using volume rendering for hollow regions of the structure and standard CPR, for the surrounding tissue. This effectively combines grey scale contextual information with detailed colour information from the area of interest. The approach is successfully used with each of the standard CPR types and the resulting images are promising as an alternative for virtual colonoscopy.

We also demonstrate how systems can effectively utilize this new visualisation in order to convey maximum information to the user. We show how overlays can be used to present surface coverage data and how sophisticated lighting models can improve the users understanding of the 3D structure. We also present details of how to integrate our visualisation into existing systems and work flows.

Publication Type:	Thesis (Doctoral)
Subjects:	R Medicine R Medicine > RZ Other systems of medicine
Departments:	School of Health & Psychological Sciences School of Health & Psychological Sciences > School of Health & Psychological Sciences Doctoral Theses Doctoral Theses

Export

Downloads