Fully automated gradient based breast boundary detection for digitized X-ray mammograms

Kus, Pelin; Karagöz, İrfan

doi:10.1016/j.compbiomed.2011.10.011

Cited by 14 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Digital breast tomosynthesis images were reconstructed into a unified spacing slice (1.0 mm) using the simultaneous algebraic reconstruction technique (SART) (Zhang et al, 2006). To save computational memory and avoid the calculation of largescale convolutions for the background pixels in the deep learning-based CAD system, the skin and the background were excluded from the breast region using a dynamic multiple thresholding-based breast boundary method (Wu et al, 2010;Kus and Karagoz, 2012).…”

Section: Image Preprocessingmentioning

confidence: 99%

Mass Detection and Segmentation in Digital Breast Tomosynthesis Using 3D-Mask Region-Based Convolutional Neural Network: A Comparative Analysis

Fan

Zheng

et al. 2020

Front. Mol. Biosci.

View full text Add to dashboard Cite

Digital breast tomosynthesis (DBT) is an emerging breast cancer screening and diagnostic modality that uses quasi-three-dimensional breast images to provide detailed assessments of the dense tissue within the breast. In this study, a framework of a 3D-Mask region-based convolutional neural network (3D-Mask RCNN) computer-aided diagnosis (CAD) system was developed for mass detection and segmentation with a comparative analysis of performance on patient subgroups with different clinicopathological characteristics. To this end, 364 samples of DBT data were used and separated into a training dataset ( n = 201) and a testing dataset ( n = 163). The detection and segmentation results were evaluated on the testing set and on subgroups of patients with different characteristics, including different age ranges, lesion sizes, histological types, lesion shapes and breast densities. The results of our 3D-Mask RCNN framework were compared with those of the 2D-Mask RCNN and Faster RCNN methods. For lesion-based mass detection, the sensitivity of 3D-Mask RCNN-based CAD was 90% with 0.8 false positives (FPs) per lesion, whereas the sensitivity of the 2D-Mask RCNN- and Faster RCNN-based CAD was 90% at 1.3 and 2.37 FPs/lesion, respectively. For breast-based mass detection, the 3D-Mask RCNN generated a sensitivity of 90% at 0.83 FPs/breast, and this framework is better than the 2D-Mask RCNN and Faster RCNN, which generated a sensitivity of 90% with 1.24 and 2.38 FPs/breast, respectively. Additionally, the 3D-Mask RCNN achieved significantly ( p < 0.05) better performance than the 2D methods on subgroups of samples with characteristics of ages ranged from 40 to 49 years, malignant tumors, spiculate and irregular masses and dense breast, respectively. Lesion segmentation using the 3D-Mask RCNN achieved an average precision (AP) of 0.934 and a false negative rate (FNR) of 0.053, which are better than those achieved by the 2D methods. The results suggest that the 3D-Mask RCNN CAD framework has advantages over 2D-based mass detection on both the whole data and subgroups with different characteristics.

show abstract

Section: Image Preprocessingmentioning

confidence: 99%

Mass Detection and Segmentation in Digital Breast Tomosynthesis Using 3D-Mask Region-Based Convolutional Neural Network: A Comparative Analysis

Fan

Zheng

et al. 2020

Front. Mol. Biosci.

View full text Add to dashboard Cite

show abstract

“…Anthropometric measurements such as these are critical to the understanding of breast shape and the development of bra products [7][8][9]. Multiple methods have been proposed to find the boundary of breasts for mammograms or for breast MR images [10][11][12][13], however, their focus is to separate breasts from the background (the ambient environment) rather than from the chest wall. Coltman, Steele and McGhee scanned their participants while they were laying in prone position, letting the breasts naturally suspend in a gap between two tables (similar to the standard posture for breast MRI) [6].…”

Section: Introductionmentioning

confidence: 99%

The Detection of the Upper Boundary of Breasts Using 4D Scanning Technology

PEI¹,

Griffin²,

FAN³

et al. 2019

Proceedings of 3DBODY.TECH 2019 - 10th International Conference and Exhibition on 3D Body Scanning and Processing Technologies,

View full text Add to dashboard Cite

Defining the boundary of female breasts is an unavoidable and very essential step before estimating the volume or surface area of the breast. Anthropometric measurements such as these are critical to the understanding of breast shape and the development of bra products. However, a non-contact method that accurately detects the boundary between breasts and chest wall, for breasts evaluated in the upright body position, has not been reported before. In this study, with the help of 4D body scanning and taking advantage of the time delay in the vertical displacement between the breasts and the chest wall during physical activity, we 4D-scanned 26 female participants using the Temporal 3dMD system and proposed a method to visualize the amount of variability for the relative displacement in the vertical direction, to facilitate the definition of the upper boundary of breasts. By viewing each breast as a cone or hemisphere, and processing the scan through vertical slicing, we were able to make all the scans having the same number of points (7200= 180×40. There were 40 vertical slices in total, and on each slice, there was one point located at every other degree from-180° to 180°) sorted in the exact same order. After proper alignment, we calculated the z-coordinate difference between a scan captured while the participant was running (there were nine dynamic scans that formed a gait cycle) and the corresponding scan of the same participant captured in static standing (the static scan). Then we were able to calculate the standard deviation (SD) value of the z-coordinate differences (i.e. the relative vertical displacements) across the nine dynamic scans. Heat maps were then created with the SD values mapped onto the 3D surface of the static scan and presented by gradient colors. This study is one of the first comprehensive studies that investigate the vertical displacement and the shape deformation of breasts during running using 4D scanning technology. Ultimately, the method and results can increase the understanding of breast kinematics, and benefit the product design of bras, especially for sports bras.

show abstract

“…The nipple location has been considered, e.g., by Yin et al (1994), Méndez et al (1996), Chandrasekhar and Attikiouzel (1997), Zhou et al (2004), Karnan and Thangavel (2007) and Kinoshita et al (2008). The breast boundary detection or the breast skin line detection has been investigated by Ojala et al (2001), Wu et al (2010) and Kus and Karagoz (2012), among others.…”

Section: Introductionmentioning

confidence: 99%

Breast tissue segmentation from x-ray radiographs

et al. 2014

View full text Add to dashboard Cite

In this paper, we propose a robust and accurate method that segments mammograms to three distinct regions: breast tissue, pectoral muscle and background. Our approach is built around a neural, two-layer committee machine. On the first layer, individual experts, each formed by a feature vector and a classifier, vote the local class label of the mammogram. The votes are given as an input, together with a prior map, to the second layer of the committee machine, which combines the inputs by a gating network. As the first layer features, we use effective, well-known local features based on image intensity, intensity histograms, local binary patterns, and histograms of oriented gradient. As with the first-layer classifiers and the gating network, we use support vector machines. Our experiments on a database of 495 mammograms, divided into independent training, validations and test subsets, show that our method is able to segment the breast tissue without failure, and it challenges the manual expert segmentation in the level of accuracy.

show abstract

Fully automated gradient based breast boundary detection for digitized X-ray mammograms

Cited by 14 publications

References 26 publications

Mass Detection and Segmentation in Digital Breast Tomosynthesis Using 3D-Mask Region-Based Convolutional Neural Network: A Comparative Analysis

Mass Detection and Segmentation in Digital Breast Tomosynthesis Using 3D-Mask Region-Based Convolutional Neural Network: A Comparative Analysis

The Detection of the Upper Boundary of Breasts Using 4D Scanning Technology

Breast tissue segmentation from x-ray radiographs

Contact Info

Product

Resources

About