Semi-automatic tumor boundary detection in MR image sequences

Abstract

The authors present a semi-automatic approach for the detection of tumor boundary in MR image sequences. An initial slice with an obvious tumor is selected from the image sequence. The tumor is roughly segmented using fuzzy c-means algorithm and its boundary can be further refined by region and contour deformation. For the rest of the slices, the initial plan applied for each slice is extracted from the resulting boundary of the previous slice. The tumor boundary is located using region and contour deformation. Performance of our approach is evaluated on the MR image sequence. Comparisons with manual tracing show the accuracy and effectiveness of our approach.