Author : Dr. Xiao Di
Last Updated : 23 June 2003

Background

Attributable to diets, bad habits, or genetic defects, colon and rectal cancer has become one of the leading diseases that damage the human health, especially in the developed countries. Colorectal cancer is more common with increasing age and incidence rates per decade double from 40 to 80 years. Both for men and women, the incidence of this disease is equal. When cancer is spread, the survival rate drops to 65% and below. Unfortunately, because many cancers are detected at later stages, the overall survival is just about 50%. So, like other cancers, the early stage diagnosis of the colorectal cancer is more important for getting good treatment.

Figure 1 Structure of ultrasound rectal wall

In the past ten years, endoscopic ultrasonography (EUS) has been gradually applied in colon and rectal screening tests and trended to more mature. By a great amount of clinical practice and with the improvement of instrument performance, endoscopic ultrasonography is now becoming a new technique to help clinical diagnosis in colorectal surgery. So ultrasound image processing and information extraction on colon and rectum has become valuable to digestive physicians (Fig. 1).

To extract useful information, surgeons need to view ultrasound images slice by slice, to delineate each layer's boundary and tumor's boundary, and finally to form a full information in their mind about tumor's status through combining results from a series of images. Then, according to modified duck's classification system or TNM staging system, surgeons can decide the staging status of rectal tumor, which expresses what area has been invaded by tumor.

At present, although some sophisticated EUS machines have been able to supply 3-dimensional ultrasound image model of rectum, it is still a time-consuming and tedious work for surgeons to perform object recognition and information extraction.

Goal of the project:

In this project, the goal is to develop an automatic analysis system to aid surgeons to cope with large quantities of rectal ultrasound images, involved in rectal layer detection, rectal wall tumor detection, 3D reconstruction of wall layer structure and corresponding quantitative analysis, which will be helpful for the early stage diagnosis of rectal cancer.  The whole project can be divided into following three subprojects:

Project 1: Multi-layer boundary detection of rectal wall

Project 2: Tumor detection on the rectal wall

Project 3: 3D rectal wall layer reconstruction

Publications related to Image Processing on Rectum.

We would be glad if you could sign our guest book.

For more information, please contact the principal investigator:

A/P Ng Wan Sing
School of Mechanical & Aerospace Engineering
Nanyang Technological University
Nanyang Avenue, Singapore 639798
Fax:(65) 6791 1859

Hospital Partner:

Dr. Charles B.S. Tsang
Consultant Surgeon, Division of Colorectal Surgery
National University Hospital
Singapore