Deep learning in mammography images segmentation and classification: Automated CNN approach

Salama, Wessam M.; Aly, Moustafa H.

doi:10.1016/j.aej.2021.03.048

Cited by 170 publications

(66 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Deep learning applied to mammography imaging has been broadly studied in past years (21-23). Recent works include deep neural networks for classification (17,18,20,24), detection (16,25), and segmentation (26)(27)(28).…”

Section: Related Workmentioning

confidence: 99%

Leveraging Multi-Task Learning to Cope With Poor and Missing Labels of Mammograms

Tardy¹,

Mateus²

2022

Front. Radio

View full text Add to dashboard Cite

In breast cancer screening, binary classification of mammograms is a common task aiming to determine whether a case is malignant or benign. A Computer-Aided Diagnosis (CADx) system based on a trainable classifier requires clean data and labels coming from a confirmed diagnosis. Unfortunately, such labels are not easy to obtain in clinical practice, since the histopathological reports of biopsy may not be available alongside mammograms, while normal cases may not have an explicit follow-up confirmation. Such ambiguities result either in reducing the number of samples eligible for training or in a label uncertainty that may decrease the performances. In this work, we maximize the number of samples for training relying on multi-task learning. We design a deep-neural-network-based classifier yielding multiple outputs in one forward pass. The predicted classes include binary malignancy, cancer probability estimation, breast density, and image laterality. Since few samples have all classes available and confirmed, we propose to introduce the uncertainty related to the classes as a per-sample weight during training. Such weighting prevents updating the network's parameters when training on uncertain or missing labels. We evaluate our approach on the public INBreast and private datasets, showing statistically significant improvements compared to baseline and independent state-of-the-art approaches. Moreover, we use mammograms from Susan G. Komen Tissue Bank for fine-tuning, further demonstrating the ability to improve the performances in our multi-task learning setup from raw clinical data. We achieved the binary classification performance of AUC = 80.46 on our private dataset and AUC = 85.23 on the INBreast dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Leveraging Multi-Task Learning to Cope With Poor and Missing Labels of Mammograms

Tardy¹,

Mateus²

2022

Front. Radio

View full text Add to dashboard Cite

show abstract

“…Furthermore, most of the reviewed papers did not mention the filters used, because it is possible to train deep learning model without any preprocessing and postprocessing modules. Moreover, the most frequently used segmentation is U-Net, and it also outperformed other models as it achieved a Dice of 98.87% [173]. (xxv) Grid Platform for Computer-Assisted Library for Mammography (GPCalma): the database consists of 3369 mammographic images [190].…”

Section: Analysis On Deepmentioning

confidence: 99%

Breast Cancer Segmentation Methods: Current Status and Future Potentials

Michael

et al. 2021

BioMed Research International

View full text Add to dashboard Cite

Early breast cancer detection is one of the most important issues that need to be addressed worldwide as it can help increase the survival rate of patients. Mammograms have been used to detect breast cancer in the early stages; if detected in the early stages, it can drastically reduce treatment costs. The detection of tumours in the breast depends on segmentation techniques. Segmentation plays a significant role in image analysis and includes detection, feature extraction, classification, and treatment. Segmentation helps physicians quantify the volume of tissue in the breast for treatment planning. In this work, we have grouped segmentation methods into three groups: classical segmentation that includes region-, threshold-, and edge-based segmentation; machine learning segmentation; and supervised and unsupervised and deep learning segmentation. The findings of our study revealed that region-based segmentation is frequently used for classical methods, and the most frequently used techniques are region growing. Further, a median filter is a robust tool for removing noise. Moreover, the MIAS database is frequently used in classical segmentation methods. Meanwhile, in machine learning segmentation, unsupervised machine learning methods are more frequently used, and U-Net is frequently used for mammogram image segmentation because it does not require many annotated images compared with other deep learning models. Furthermore, reviewed papers revealed that it is possible to train a deep learning model without performing any preprocessing or postprocessing and also showed that the U-Net model is frequently used for mammogram segmentation. The U-Net model is frequently used because it does not require many annotated images and also because of the presence of high-performance GPU computing, which makes it easy to train networks with more layers. Additionally, we identified mammograms and utilised widely used databases, wherein 3 and 28 are public and private databases, respectively.

show abstract

“…ResNet50 achieves an accuracy of 96% [19]. Salama and Aly [20] obtained an accuracy of 98.87% with the U-Net model and InceptionV3 with the data augmentation for the breast cancer classification. Data augmentation influences the average precision of the class [21][22][23].…”

Section: Related Workmentioning

confidence: 99%

Effect of Data Augmentation in the Classification and Validation of Tomato Plant Disease with Deep Learning Methods

Wagle¹,

Sampe²,

Faseehuddin³

et al. 2021

View full text Add to dashboard Cite

The paper discusses disease identification and classification in tomato plants, as well as the effect of data augmentation in deep learning models. The database used here is Tomato plant leaves (TPL) images from the PlantVillage Database in the healthy and disease classes. The disease categories have been chosen depending on their occurrence in the Indian States. The proposed ResNet50, ResNet18, and ResNet101 deep-learning model with transfer learning combined with the softmax classification are used to identify and categorize the tomato leaf images into the healthy or diseases classes in the dataset. The unique combination of including the noise and blur in the images and position and color data augmentation makes the dataset robust. Two different data augmentation methods are used for the classification problem, and significant improvement is seen in the classification accuracy with the proposed augmented dataset. The model’s success rate makes the model helpful in extending support in validating a model for identifying plant disease. The validation of models is done on PlantVillage and images taken at Krishi Vigyan Kendra Narayangaon, Pune, India. ResNet101 model trained with augmented dataset outperforms the testing accuracy of 99.99% and validation accuracy of 95.83%.

show abstract

Deep learning in mammography images segmentation and classification: Automated CNN approach

Cited by 170 publications

References 27 publications

Leveraging Multi-Task Learning to Cope With Poor and Missing Labels of Mammograms

Leveraging Multi-Task Learning to Cope With Poor and Missing Labels of Mammograms

Breast Cancer Segmentation Methods: Current Status and Future Potentials

Effect of Data Augmentation in the Classification and Validation of Tomato Plant Disease with Deep Learning Methods

Contact Info

Product

Resources

About