Malignant and nonmalignant classification of breast lesions in mammograms using convolutional neural networks

Abstract Early detection remains the backbone of breast cancer control and treatment improvement. However, early detection is difficult since cancer symptoms are absent at onset. Hence, cancer remains one of the health topics that many researchers try to improve from diagnosis, prevention, and treatment perspectives. The main aim of this research is to develop a system using deep learning to classify breast lesions in mammographic images into malignant and nonmalignant based on two approaches, one by using patches of region of interest (ROI) and the other using the whole images. The proposed system comprises two phases, which are the preprocessing phase and the Convolution Neural Network (CNN) building phase. The preprocessing phase prepares the mammogram images for the CNN building phase, it includes format unification, noise removal, image enhancement, ROI extraction, augmentation, and image resizing. The CNN building phase builds the proposed CNN model from scratch to learn features and classify the breast lesions in mammogram images. The proposed system offers good classification rates. It was applied on three benchmark datasets, which are the MIAS, DDSM, and INbreast. Using 5-fold cross validation, experimental results demonstrated high performance and promising results. The sensitivity, specificity, accuracy and AUC reached 96.55%, 96.49%, 96.52%, and 0.98 respectively for the INbreast dataset, and reached 98%, 92.6%, 95.3%, and 0.974 respectively for the MIAS dataset.

[1]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[2]  Syed Muhammad Anwar,et al.  Medical Image Analysis using Convolutional Neural Networks: A Review , 2017, Journal of Medical Systems.

[3]  Ming Yang,et al.  Classification of Alzheimer’s Disease Based on Eight-Layer Convolutional Neural Network with Leaky Rectified Linear Unit and Max Pooling , 2018, Journal of Medical Systems.

[4]  Jaime S. Cardoso,et al.  INbreast: toward a full-field digital mammographic database. , 2012, Academic radiology.

[5]  Zhenwei Zhang,et al.  Radiological images and machine learning: trends, perspectives, and prospects , 2019, Comput. Biol. Medicine.

[6]  Miguel López-Coronado,et al.  Predicting Absenteeism and Temporary Disability Using Machine Learning: a Systematic Review and Analysis , 2020, Journal of Medical Systems.

[7]  Banshidhar Majhi,et al.  Automated breast cancer detection in digital mammograms: A moth flame optimization based ELM approach , 2020, Biomed. Signal Process. Control..

[8]  Lazaros T. Tsochatzidis,et al.  Computer-aided diagnosis of mammographic masses based on a supervised content-based image retrieval approach , 2017, Pattern Recognit..

[9]  Z. Jane Wang,et al.  A Computer-Aided Decision Support System for Detection and Localization of Cutaneous Vasculature in Dermoscopy Images Via Deep Feature Learning , 2018, Journal of Medical Systems.

[10]  Nisreen I. R. Yassin,et al.  Machine learning techniques for breast cancer computer aided diagnosis using different image modalities: A systematic review , 2018, Comput. Methods Programs Biomed..

[11]  Abhishek Midya,et al.  Neighborhood Structural Similarity Mapping for the Classification of Masses in Mammograms , 2018, IEEE Journal of Biomedical and Health Informatics.

[12]  István Csabai,et al.  Detecting and classifying lesions in mammograms with Deep Learning , 2017, Scientific Reports.

[13]  Quan Liu,et al.  Electrocardiogram generation with a bidirectional LSTM-CNN generative adversarial network , 2019, Scientific Reports.

[14]  Chuanbo Guo,et al.  Impacts of hatchery-reared mandarin fish Siniperca chuatsi stocking on wild fish community and water quality in a shallow Yangtze lake , 2018, Scientific Reports.

[15]  Bodhisattva Dash,et al.  Automated diagnosis of breast cancer using parameter optimized kernel extreme learning machine , 2020, Biomed. Signal Process. Control..

[16]  Allan Kardec Barros,et al.  Detecting masses in dense breast using independent component analysis , 2017, Artif. Intell. Medicine.

[17]  J. Dinesh Peter,et al.  Classification of Mammogram Images Using Multiscale all Convolutional Neural Network (MA-CNN) , 2019, Journal of Medical Systems.

[18]  F. F. Ting,et al.  Breast cancer detection using convolutional neural networks for mammogram imaging system , 2017, 2017 International Conference on Robotics, Automation and Sciences (ICORAS).

[19]  Hua Li,et al.  Benign and malignant classification of mammogram images based on deep learning , 2019, Biomed. Signal Process. Control..

[20]  Marcin Lewandowski,et al.  Breast lesion classification based on ultrasonic radio-frequency signals using convolutional neural networks , 2020, Biocybernetics and Biomedical Engineering.

[21]  Gustavo Carneiro,et al.  A deep learning approach for the analysis of masses in mammograms with minimal user intervention , 2017, Medical Image Anal..

[22]  Li Shen,et al.  Deep Learning to Improve Breast Cancer Detection on Screening Mammography , 2017, Scientific Reports.

[23]  Ji Wan,et al.  Deep Learning for Content-Based Image Retrieval: A Comprehensive Study , 2014, ACM Multimedia.

[24]  Chao Wang,et al.  Optimal breast tumor diagnosis using discrete wavelet transform and deep belief network based on improved sunflower optimization method , 2020, Biomed. Signal Process. Control..

[25]  Justin Joseph,et al.  A fully customized enhancement scheme for controlling brightness error and contrast in magnetic resonance images , 2018, Biomed. Signal Process. Control..