Use of a variable thresholding-based image segmentation technique for magnetic resonance guided High Intensity Focused Ultrasound therapy: An in vivo validation

Abstract

In this paper, the implementation of a segmentation technique based on Otsu's method, region growing algorithm and selection of regions using global and variable thresholding for the treatment planning of Magnetic Resonance guided High Intensity Focused Ultrasound (MRgHIFU) is described. The method is used to classify the pixels of real Magnetic Resonance (MR) images obtained for the study of the distribution of heat in abscess treatment in a murine model with High-Intensity Focused Ultrasound (HIFU). Using a discrepancy measure for evaluation, the proposed technique (segmentation technique I) demonstrated to be more efficient than an approach technique (segmentation technique II) based on Otsu's method, global thresholding, edge detection and region growing algorithm. In the evaluation, a total of nine surveys of 48 images each were used. For axial images the performance of segmentation technique I and the performance of segmentation technique II is very similar, having an average value of 92.05% for the former and an average value of 91.45% for the latter. On the other hand, for sagittal images, segmentation technique I presented an average performance of 85.46% while segmentation technique II presented an average performance of 69.01%.