Deep Generative Breast Cancer Screening and Diagnosis

Shams, Soheil; Platania, Richard; Zhang, Jian; Kim, Joohyun; Lee, Kisung; Park, Seung-Jong

doi:10.1007/978-3-030-00934-2_95

Cited by 50 publications

(34 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It shows that, the Diag-Net has achieved the state-of-art with mean accuracy 93.4% and AUC score 0.95. When compared with the second best algorithm [16], the DiagNet's AUC score is significantly higher with experiments on the whole dataset without any pre-processing, post-processing or transfer learning. In addition, empirical observations show that our model is robust to noise and geometric transforms, and these results are omitted due to the space limitation.…”

Section: Results and Analysismentioning

confidence: 94%

“…As also mentioned in section 1, inadequate data and the similarity between benign and cancerous masses [7] are two main reasons causing high false positives in mammographic CADs. Recently, [1,16,17] employed GANs to create new instances. Even though they generated on-distribution samples that are not separable by discriminators, they ignored the importance of distinguishable but similar instances, which tend to improve the discriminative ability.…”

Section: Adversarial Augmentationmentioning

confidence: 99%

“…Mammography is extensively applied and computer-aided diagnosis systems (CADs) are often employed as a second reader. Leveraging the recent success of deep neural networks on representation learning, deep learning based CADs [7,11,12,16,17,20] outperform traditional methods, which rely heavily on handcrafted features. However, two major challenges in mammographic CADs remain (1): limited access to well annotated data [7] and (2): the similarity between benign and cancerous masses.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Signed Laplacian Deep Learning with Adversarial Augmentation for Improved Mammography Diagnosis

Chen

Nailon

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Computer-aided breast cancer diagnosis in mammography is limited by inadequate data and the similarity between benign and cancerous masses. To address this, we propose a signed graph regularized deep neural network with adversarial augmentation, named DiagNet. Firstly, we use adversarial learning to generate positive and negative mass-contained mammograms for each mass class. After that, a signed similarity graph is built upon the expanded data to further highlight the discrimination. Finally, a deep convolutional neural network is trained by jointly optimizing the signed graph regularization and classification loss. Experiments show that the DiagNet framework outperforms the state-of-the-art in breast mass diagnosis in mammography.

show abstract

Section: Results and Analysismentioning

confidence: 94%

Section: Adversarial Augmentationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Signed Laplacian Deep Learning with Adversarial Augmentation for Improved Mammography Diagnosis

Chen

Nailon

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Traditional mammographic computer-aided diagnosis (CAD) systems rely heavily on elaborate hand-crafted fea- tures [2]. Recently, leveraging the insights from the successes of deep neural networks (deep learning) in computer vision tasks [5][6][7][8][9], deep learning based algorithms have been applied to mammograms and have achieved state-of-the-art results for mass detection, segmentation and classification. Mass detection aims to find the regions of interest (ROIs) where abnormalities may be located, and mass segmentation provides detailed morphological features with precise outlines within ROIs.…”

Section: Introductionmentioning

confidence: 99%

“…To address this, a novel architecture is designed in this paper. Based on the accurate pixel-level labelling algorithm presented in [4] and the related breast mass classifiers [2,3,5,7,8,10,11], a Dual-path Conditional Residual Network (DUALCORENET) for mammography analysis is proposed as shown in Figure 1. Firstly, a mass and its context texture learner called the Locality Preserving Learner (LPL) is built with stacks of convolutional blocks, achieving a mapping from ROIs to class labels.…”

Section: Introductionmentioning

confidence: 99%

A Deep Dual-path Network for Improved Mammogram Image Processing

Chen

Nailon

et al. 2019

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We present, for the first time, a novel deep neural network architecture called DUALCORENET with a dual-path connection between the input image and output class label for mammogram image processing. This architecture is built upon U-Net, which non-linearly maps the input data into a deep latent space. One path of the DUALCORENET, the locality preserving learner, is devoted to hierarchically extracting and exploiting intrinsic features of the input, while the other path, called the conditional graph learner, focuses on modeling the input-mask correlations. The learned mask is further used to improve classification results, and the two learning paths complement each other. By integrating the two learners our new architecture provides a simple but effective way to jointly learn the segmentation and predict the class label. Benefiting from the powerful expressive capacity of deep neural networks a more discriminative representation can be learned, in which both the semantics and structure are well preserved. Experimental results show that DUALCORENET achieves the best mammography segmentation and classification simultaneously, outperforming recent state-of-the-art models.

show abstract

Malignancy detection on mammograms by integrating modified convolutional neural network classifier and texture features

Melekoodappattu¹,

Dhas²,

Kumar³

et al. 2021

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Breast cancer is detected by identifying malignancy on breast tissue. Emerging technologies in medical image processing are used to interpret histopathology images. For analyzing medical imaging and pathological data, modified deep neural networks are being used. Automatic detection of malignancy is usually achieved in deep learning by capturing features from a convolutional neural network (CNN) and then categorizing them using a fully connected network. A framework to automatically diagnose malignancy using an ensemble approach, including CNN and extraction of image texture features, is implemented in this research. In the CNN phase, the nine‐layer modified CNN is used to classify images. Texture features are derived and their dimension is minimized using maximum variance unfolding to enhance the efficiency of classification in the extraction‐based phase. The results of each phase were then merged to obtain the final decision. The testing specificity and accuracy of our ensemble method on MIAS repository are 98.9% and 99%, respectively and for the DDSM repository are 98.3% and 98.1%. The ensemble approach increases the measurement metrics compared to each phase separately, as per the experimental findings.

show abstract

Deep Generative Breast Cancer Screening and Diagnosis

Cited by 50 publications

References 12 publications

Signed Laplacian Deep Learning with Adversarial Augmentation for Improved Mammography Diagnosis

Signed Laplacian Deep Learning with Adversarial Augmentation for Improved Mammography Diagnosis

A Deep Dual-path Network for Improved Mammogram Image Processing

Malignancy detection on mammograms by integrating modified convolutional neural network classifier and texture features

Contact Info

Product

Resources

About