Segmentation of malignant lesions in 3D breast ultrasound using a depth‐dependent model

Tan, Tao; Gubern-Mérida, Albert; Borelli, Cristina; Manniesing, Rashindra; Zelst, Jan van; Wang, Lei; Zhang, Wei; Platel, Bram; Mann, Ritse M.; Karssemeijer, Nico

doi:10.1118/1.4953206

Cited by 13 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The segmentation can be controlled by setting the weights (α, ε) of propagation, curvature and advection term. The propagation term controls the inflation or 'balloon force' of the segmentation, the curvature term controls the 'smoothness' of the boundary of the segmentation and the advection term in the update equation attracts the contour to the lesion edge [14].…”

Section: The Level Set Methodsmentioning

confidence: 99%

“…Their best intersection of the computer and the reference segmented area was 0.84. Tan et al [14] proposed a novel depth-dependent dynamic programming technique and obtained a Dice similarity coefficient (DSC) of 0.73. This accuracy was then improved by Kozegar et al with a specific level set algorithm [15].…”

Section: Fig 1 a Malignant Lesion In Breast Ultrasoundmentioning

confidence: 99%

See 1 more Smart Citation

Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Xing

Wang

et al. 2021

IEEE/ACM Trans. Comput. Biol. and Bioinf.

Self Cite

View full text Add to dashboard Cite

Breast cancer is the most common invasive cancer with the highest cancer occurrence in females. Handheld ultrasound is one of the most efficient ways to identify and diagnose the breast cancer. The area and the shape information of a lesion is very helpful for clinicians to make diagnostic decisions. In this study we propose a new deep-learning scheme, semi-pixel-wise cycle generative adversarial net (SPCGAN) for segmenting the lesion in 2D ultrasound. The method takes the advantage of a fully convolutional neural network (FCN) and a generative adversarial net to segment a lesion by using prior knowledge. We compared the proposed method to a fully connected neural network and the level set segmentation method on a test dataset consisting of 32 malignant lesions and 109 benign lesions. Our proposed method achieved a Dice similarity coefficient (DSC) of 0.92 while FCN and the level set achieved 0.90 and 0.79 respectively. Particularly, for malignant lesions, our method increases the DSC (0.90) of the fully connected neural network to 0.93 significantly (p <0.001). The results show that our SPCGAN can obtain robust segmentation results. The framework of SPCGAN is particularly effective when sufficient training samples are not available compared to FCN. Our proposed method may be used to relieve the radiologists' burden for annotation.

show abstract

Section: The Level Set Methodsmentioning

confidence: 99%

Section: Fig 1 a Malignant Lesion In Breast Ultrasoundmentioning

confidence: 99%

Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Xing

Wang

et al. 2021

IEEE/ACM Trans. Comput. Biol. and Bioinf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…. ) and was extended for malignant lesion using a depth-guided dynamic programming method 23 ( DSC = ± 0 73 0 14 . .…”

Section: Introductionmentioning

confidence: 99%

Lesion Segmentation in Automated 3D Breast Ultrasound: Volumetric Analysis

et al. 2017

Self Cite

View full text Add to dashboard Cite

Mammography is the gold standard screening technique in breast cancer, but it has some limitations for women with dense breasts. In such cases, sonography is usually recommended as an additional imaging technique. A traditional sonogram produces a two-dimensional (2D) visualization of the breast and is highly operator dependent. Automated breast ultrasound (ABUS) has also been proposed to produce a full 3D scan of the breast automatically with reduced operator dependency, facilitating double reading and comparison with past exams. When using ABUS, lesion segmentation and tracking changes over time are challenging tasks, as the three-dimensional (3D) nature of the images makes the analysis difficult and tedious for radiologists. The goal of this work is to develop a semi-automatic framework for breast lesion segmentation in ABUS volumes which is based on the Watershed algorithm. The effect of different de-noising methods on segmentation is studied showing a significant impact ([Formula: see text]) on the performance using a dataset of 28 temporal pairs resulting in a total of 56 ABUS volumes. The volumetric analysis is also used to evaluate the performance of the developed framework. A mean Dice Similarity Coefficient of [Formula: see text] with a mean False Positive ratio [Formula: see text] has been obtained. The Pearson correlation coefficient between the segmented volumes and the corresponding ground truth volumes is [Formula: see text] ([Formula: see text]). Similar analysis, performed on 28 temporal (prior and current) pairs, resulted in a good correlation coefficient [Formula: see text] ([Formula: see text]) for prior and [Formula: see text] ([Formula: see text]) for current cases. The developed framework showed prospects to help radiologists to perform an assessment of ABUS lesion volumes, as well as to quantify volumetric changes during lesions diagnosis and follow-up.

show abstract

“…However, as a safe and inexpensive imaging modality, ultrasound is the most commonly used modality in practice. Therefore, an automatic diagnostic assistance system for ultrasound images can provide objective diagnostic support for physicians to discriminate the renal lesions and consequently improve the detection of diseases as well as eliminate manual procedures …”

Section: Introductionmentioning

confidence: 99%

“…Therefore, an automatic diagnostic assistance system for ultrasound images can provide objective diagnostic support for physicians to discriminate the renal lesions and consequently improve the detection of diseases as well as eliminate manual procedures. 6,17 In this proposed method, two bottom-up saliency detection models based on global information are implemented to endow saliency values for every pixel, the potential lesion regions are extracted as distinct regions from these two saliency maps. In the subsequent segmentation, local information, i.e., local intensity information and a novel DoGÀLBP feature in local areas, and edge indicators which are represented by intensity variance in small patches are used in the vector valued Chan-Vese (CV) model for the fine segmentation of lesions.…”

Section: Introductionmentioning

confidence: 99%

Automatic renal lesion segmentation in ultrasound images based on saliency features, improved LBP, and an edge indicator under level set framework

Gui

Yang

2017

Medical Physics

View full text Add to dashboard Cite

Purpose: Segmentation of lesions in ultrasound images is widely used for preliminary diagnosis. In this paper, we develop an automatic segmentation algorithm for multiple types of lesions in ultrasound images. The proposed method is able to detect and segment lesions automatically as well as generate accurate segmentation results for lesion regions. Methods: In the detection step, two saliency detection frameworks which adopt global image information are designed to capture the differences between normal and abnormal organs as well as these between lesions and the normal tissues around them. In the segmentation step, three types of local information, i.e., image intensity, improved local binary patterns (LBP) features, and an edge indicator, are embedded into a modified level set framework to carry out the segmentation task. Results: The cyst and carcinoma regions in the ultrasound images of the human kidneys can be automatically detected and segmented by using the proposed method. The efficiency and accuracy of the method are validated by quantitative evaluations and comparative measurements with three wellrecognized segmentation methods. Specifically, the average precision and dice coefficient of the proposed method in segmenting renal cysts are 95.33% and 90.16%, respectively, while those in segmenting renal carcinomas are 94.22% and 91.13%, respectively. The average precision and dice coefficient of the proposed method are higher than those of three compared segmentation methods. Conclusions: The proposed method can efficiently detect and segment the renal lesions in ultrasound images. In addition, since the proposed method utilizes the differences between normal and abnormal organs as well as these between lesions and the normal tissues around them, it can be possibly extended to deal with lesions in other organs of ultrasound images as well as lesions in medical images of other modalities.

show abstract

Segmentation of malignant lesions in 3D breast ultrasound using a depth‐dependent model

Abstract: The proposed depth-guided dynamic programming method achieves accurate breast malignant lesion segmentation results in automated breast ultrasound.

Cited by 13 publications

References 30 publications

Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Lesion Segmentation in Automated 3D Breast Ultrasound: Volumetric Analysis

Automatic renal lesion segmentation in ultrasound images based on saliency features, improved LBP, and an edge indicator under level set framework

Contact Info

Product

Resources

About