Image analysis for patient management in colorectal cancer

Colorectal cancer is a major health issue in the western world and accurate staging of this disease is crucial for informing patient management decisions, particularly for the surgically complex tumors. Current methods used by clinicians are primarily subjective and do not make use of much of the information potentially available from MR images. We are developing a series of image analysis methods to aid clinicians: by removing MR artifacts, segmenting and visualizing the data in three dimensions, and automatically detecting the lymph nodes and predicting whether or not they are involved in malignancy. This information is then displayed in a way that is useful to the clinicians as they assess the best course of treatment. © 2005 CARS and Elsevier B.V.