Automated lesion segmentation of clinical imaging studies is of potential value for treatment monitoring and radiation treatment planning. With low spatial resolution imaging systems, such as MR Spectroscopic Imaging, segmentation based on image intensity variations must take into consideration the broad spatial response function. In addition, the relative lesion-to-background intensity variation and the object size must be considered. In this report a new automated image segmentation method is presented that accounts for these factors, which is based on a subtraction of a smoothed version of the MRSI maps from the original data.
This abstract and the presentation materials are available to members only; a login is required.