Automated Breast Cancer Detection and Classification in Full Field Digital Mammograms Using Two Full and Cropped Detection Paths Approach

Abstract

Breast cancer is one of the most severe diseases that threaten women's life results in increasing the death rate annually as confirmed by the World Health Organization. Breast cancer early detection is one of the main reasons behind reducing cancer severity. However, with the huge number of mammograms taken daily, the checking process conducted by radiologists becomes lengthy, tiring, and pruning to errors process. Hence, with the tremendous success achieved by utilizing CNNs in bioinformatics, the development of Computer-Aided Detection (CAD) systems has proved its necessity to solve the challenging cases for the biopsies missed by the ordinary checking leads to decreasing the false positive and negative rates. In this paper, we present a YOLOV4 based CAD system to localize lesions in full and cropped mammograms and then classify them to obtain their pathology type. The proposed method mainly consists of three phases that are applied on the full-field digital mammograms of the INbreast dataset. First, the mammograms are preprocessed to remove any extra artifacts and then cropped into small, overlapped slices. Second, masses are localized through two paths: the full mammograms and the cropped slices detection after configuring the YOLO-V4 model. Third, other feature extractors like ResNet, VGG, Inception, etc. are used to classify the localized lesions to compare their performance against YOLO. The proposed method proved using the experimental results the impact of utilizing YOLO-V4 as a detector with the 2-paths of detection of a full mammogram and the cropped slices in a trial to avoid any data loss by resizing the large-sized mammograms. Our system succeeds in detecting the masses' location with an overall accuracy of ≈98% which is more than the recently introduced breast cancer detection methods. Moreover, its ability to distinguish between benign and malignant tumors with an accuracy of ≈95%.

Breast Mammogram Analysis and Classification Using Deep Convolution Neural Network

“…From the overall comparison, the average accuracy obtained by the proposed method is 99.90%, which is increased than other methods. Also, the accuracy of the proposed approach has been compared to four contemporary works recently carried out [31][32][33][34] and the results indicate that the proposed approach yields a higher classification accuracy of 99% when compared to the recent works. The advantages of the proposed architecture-4 model are obtained from Fig.…”

“…Very recently a multi-scale attention network with hierarchical block-wise and layer-wise feature representation capability [31] has been proposed. Also, A YOLOV4 based CAD system [32] to localize lesions in full and cropped mammograms was developed to classify lesions to obtain their pathology type. Another approach based on Generative Adversarial Networks (GAN) and bilateral asymmetry [33] has been used for detection of breast cancer.…”

Breast Mass Detection and Visualization with Complementary Deep Learning Architectures

“…Object-detection systems identify suspicious lesions with high precision and closely conforming boundaries, but are prone to false positives and false negatives. Numerous studies based on the YOLO architecture [25,[30][31][32][33] and other object-detection models [34][35][36][37][38], for example, have reported detection rates for mammogram masses ranging from 0.74 to over 0.98. We easily obtained a detection fraction of 0.94 using our YOLO stage on the CBIS-DDSM test image set by reducing the con dence threshold from 0.3 to 0.075; but false positives soared from 62 to 238, reducing precision from 0.74 to 0.18 -essentially an invitation to hunt for the true needle in a haystack of erroneous detections.…”

A Hybrid Workflow of Residual Convolutional Transformer Encoder for Breast Cancer Classification Using Digital X-ray Mammograms

“…A collection of investigations was undertaken to identify benign from malignant patients using mammography images. The deep learning YOLO predictor was applied to distinguish benign from malignant cases on mammogram images, as in the previous studies [ 9 , 10 , 11 , 12 , 13 ]. Al-Antari et al [ 9 ] proposed a deep learning recognition framework based on the YOLO predictor for breast images detection and classification to distinguish benign from malignant cases.…”

“…They achieved the overall classification performance of 89.5% of the overall accuracy. Hamed et al [ 11 ] utilized the YOLOV4-based CAD system to recognize the benign cases from malignant ones too, while different feature extractors such as Inception, ResNet, VGG were employed to classify the localized lesions to benign or malignant cases. The proposed model based on YOLO-V4 model outperformed other classifiers and achieved an accuracy of 98% for detecting the location of masses, while the best classification accuracy achieved by the ResNet was 95%.…”