Computer Vision-Based Microcalcification Detection in Digital Mammograms Using Fully Connected Depthwise Separable Convolutional Neural Network

Abstract

Microcalcification clusters in mammograms are one of the major signs of breast cancer. However, the detection of microcalcifications from mammograms is a challenging task for radiologists due to their tiny size and scattered location inside a denser breast composition. Automatic CAD systems need to predict breast cancer at the early stages to support clinical work. The intercluster gap, noise between individual MCs, and individual object’s location can affect the classification performance, which may reduce the true-positive rate. In this study, we propose a computer-vision-based FC-DSCNN CAD system for the detection of microcalcification clusters from mammograms and classification into malignant and benign classes. The computer vision method automatically controls the noise and background color contrast and directly detects the MC object from mammograms, which increases the classification performance of the neural network. The breast cancer classification framework has four steps: image preprocessing and augmentation, RGB to grayscale channel transformation, microcalcification region segmentation, and MC ROI classification using FC-DSCNN to predict malignant and benign cases. The proposed method was evaluated on 3568 DDSM and 2885 PINUM mammogram images with automatic feature extraction, obtaining a score of 0.97 with a 2.35 and 0.99 true-positive ratio with 2.45 false positives per image, respectively. Experimental results demonstrated that the performance of the proposed method remains higher than the traditional and previous approaches.

A Newly Hybrid Version of African Vultures Optimization Algorithm and Sine Cosine Algorithm for Mammographic Masses and its Diagnosis

“…Khalil ur Rehman et. al [10] gave a model for classifying the microcalcification in the mammograms using FC-DSCNN (Fully Connected Depthwise Separable Convolutional Neural Network). The experimentation is done on a dataset collected locally from a hospital approved by Diagnostic Imaging Nuclear Medicine and Radiology Research and Development Committee consisting of total 577 patients.…”

“…For assessing the hunting behaviour of the AVSC method has been considered unimodal functions which involve a single optima value. Tabulated results of optimizers on these functions have been reported by tables (5)(6)(7)(8)(9)(10). Attained solutions reveal that the AVSC method is able to trap the best optima value in the complex domain with least number of iterations than others.…”

“…Consequently, these undesirable radiopaque artifacts are impassive by modest thresholding procedure and the process of the similar as well as the breast outline subdivision is illustrated in fig. (10).…”

An Automated In-Depth Feature Learning Algorithm for Breast Abnormality Prognosis and Robust Characterization from Mammography Images Using Deep Transfer Learning

“…The most widely used deep learning technique is CNN, which enables automatic mass recognition, feature learning and classification, applying smaller training datasets without human intervention. CNNs are constructed as a layer hierarchy [ 10 ]. Each layer converts input images to abstract images composed of edges, noise, and objects, and the final layer performs predictions using the pooled features [ 11 ].…”

State-of-the-Art Challenges and Perspectives in Multi-Organ Cancer Diagnosis via Deep Learning-Based Methods

“…Due to narrow size and fatty tissue problems, the early detection of malignant tumors is challenging. The timely detection of breast cancer can help with the diagnosis procedure, which can alleviate the disease severity with more excellent recovery [ 18 , 19 ]. Therefore, state-of-the-art, fully automatic methods are needed for early breast tumor detection.…”