Automatic Segmentation of Ultrasound Tomography Image

Abstract

Ultrasound tomography (UST) image segmentation is fundamental in breast density estimation, medicine response analysis, and anatomical change quantification. Existing methods are time consuming and require massive manual interaction. To address these issues, an automatic algorithm based on GrabCut (AUGC) is proposed in this paper. The presented method designs automated GrabCut initialization for incomplete labeling and is sped up with multicore parallel programming. To verify performance, AUGC is applied to segment thirty-two in vivo UST volumetric images. The performance of AUGC is validated with breast overlapping metrics (Dice coefficient (D), Jaccard (J), and False positive (FP)) and time cost (TC). Furthermore, AUGC is compared to other methods, including Confidence Connected Region Growing (CCRG), watershed, and Active Contour based Curve Delineation (ACCD). Experimental results indicate that AUGC achieves the highest accuracy (D = 0.9275 and J = 0.8660 and FP = 0.0077) and takes on average about 4 seconds to process a volumetric image. It was said that AUGC benefits large-scale studies by using UST images for breast cancer screening and pathological quantification.

Update on thyroid ultrasound

“…[41] In general, quantitative metrics, namely, Dice coefficient, Jaccard coefficient, Boundary displacement error, and global consistency error are adopted to validate the segmentation. [42,43] A quantitative analysis is convenient for the horizontal comparison of segmentation efficiency using different algorithm methods.…”

Automated Microwave Tomography (MWT) Image Segmentation: State-of-the-Art Implementation and Evaluation

“…Likewise, Dorgham [1] deployed an automatic segmentation method on the basis of GrabCut [17] to detect human body Regions of Interest (RoI) from CT images. Wu et al [18] also proposed an automatic segmentation algorithm named AUGC in ultrasound tomography images for breast cancer screening and pathological quantification. This method was also established on GrabCut as well, addressing incomplete labeling and speeding up multicore parallel programming.…”

“…• Jaccard index: This is a commonly used metric to derive the similarity and diversity of finite sample sets. We choose this statistic to gauge our study as it has been extensively applied in the context of image segmentation [3,18]. In our paper, it indicates the pixel-level similarity of input and output image.…”

“…The input MWT image is referred as ground-truth while the output image represents the segmented image. The mathematical form of the Jaccard index is defined as (In this article, G and S annotate the input ground-truth image and the output segmented image respectively amid this denotation [18]. ):…”

Erratum to “Automatic Segmentation of Ultrasound Tomography Image”

“…In the article titled “Automatic Segmentation of Ultrasound Tomography Image” [ 1 ], the affiliation of the third author was incorrect. The corrected authors' list and affiliations are shown above.…”