Efficient Breast Cancer Diagnosis from Complex Mammographic Images Using Deep Convolutional Neural Network

Abstract

Medical image analysis places a significant focus on breast cancer, which poses a significant threat to women’s health and contributes to many fatalities. An early and precise diagnosis of breast cancer through digital mammograms can significantly improve the accuracy of disease detection. Computer-aided diagnosis (CAD) systems must analyze the medical imagery and perform detection, segmentation, and classification processes to assist radiologists with accurately detecting breast lesions. However, early-stage mammography cancer detection is certainly difficult. The deep convolutional neural network has demonstrated exceptional results and is considered a highly effective tool in the field. This study proposes a computational framework for diagnosing breast cancer using a ResNet-50 convolutional neural network to classify mammogram images. To train and classify the INbreast dataset into benign or malignant categories, the framework utilizes transfer learning from the pretrained ResNet-50 CNN on ImageNet. The results revealed that the proposed framework achieved an outstanding classification accuracy of 93%, surpassing other models trained on the same dataset. This novel approach facilitates early diagnosis and classification of malignant and benign breast cancer, potentially saving lives and resources. These outcomes highlight that deep convolutional neural network algorithms can be trained to achieve highly accurate results in various mammograms, along with the capacity to enhance medical tools by reducing the error rate in screening mammograms.

BreastCDNet: Breast Cancer Detection Neural Network, Classification and Localization

“…The use of EfficientNet as the base model enhances system performance compared to previous architectures. 15 Weighted Residual Connections 16 Cross-Stage Partial 17 Cross Mini-Batch Normalization 18 Self-Adversarial Training 19 Complete Intersection over union 20 Average Precision 21 Company Owned Company Operated 22 Extreme gradient boosting 23 Computer Aided Detection Rahman et al [39] proposed a system focuses on detecting malignant breast masses and accurately classifying benign and malignant tissues in mammograms, that incorporates thresholding and region-based segmentation techniques. The region-based method employs a threshold of 80 to identify the largest area within this threshold.…”

A Systematic Literature Review of 3D Deep Learning Techniques in Computed Tomography Reconstruction

“…CNNs [101] analyze the input data in the context of computed tomography reconstruction by applying several trainable filters that convolve across the input's structural dimensions. Local patterns and characteristics may be extracted from the computed tomography data using this convolution procedure [102,103]. CNNs may learn representations that are especially suited to added tomography reconstruction tasks by stacking many convolutional layers, which allows them to capture increasingly complicated and abstract aspects from the input data.…”

Diagnosis of Medical Images Using Convolutional Neural Networks

“…However, interpreting medical images can be difficult and time-consuming, often requiring the expertise of radiologists and other medical specialists. The introduction of Convolutional Neural Networks (CNNs) has transformed how medical pictures are evaluated and interpreted [2]. CNNs, a type of deep learning algorithm inspired by the human brain's visual cortex, have shown an extraordinary ability to automatically learn and extract relevant characteristics from images [3].…”