Breast Cancer Classification Using FCN and Beta Wavelet Autoencoder

Abstract

In this paper, a new classification approach of breast cancer based on Fully Convolutional Networks (FCNs) and Beta Wavelet Autoencoder (BWAE) is presented. FCN, as a powerful image segmentation model, is used to extract the relevant information from mammography images. It will identify the relevant zones to model while WAE is used to model the extracted information for these zones. In fact, WAE has proven its superiority to the majority of the features extraction approaches. The fusion of these two techniques have improved the feature extraction phase and this by keeping and modeling only the relevant and useful features for the identification and description of breast masses. The experimental results showed the effectiveness of our proposed method which has given very encouraging results in comparison with the states of the art approaches on the same mammographic image base. A precision rate of 94% for benign and 93% for malignant was achieved with a recall rate of 92% for benign and 95% for malignant. For the normal case, we were able to reach a rate of 100%.

Application of Deep Learning in Histopathology Images of Breast Cancer: A Review

“…AlexNet (Figure 2C) is a deep convolutional neural network proposed by Hinton et al, in 2012 and won the championship in the ImageNet challenge that year. Figure 2D is a full convolution neural network (FCN) used for semantic segmentation in the early stage [24]. It is also widely used in the early research of breast cancer pathological image segmentation task.…”

Investigation on Effects of Training Schemes and Data Characteristics on Deep Learning-based Breast Cancer Classification

“…Autoencoder and CNN models are used by Togakar et al [14] to classify invasive ductal carcinoma breast cancer with an accuracy of 98.59%. Aleisa et al [15] present a novel method for classifying breast cancer with a recall rate of 92% (for benign),95% (for malignant tumor) using fully convolutional networks and beta wavelet autoencoders. A fresh approach has been proposed by Amin et al [16]to diagnose breast cancer.…”

Breast Cancer Classification Utilizing Deep Learning Techniques on Medical Images: A Comprehensive Review

“…The autoencoder performs better than its rivals, with a 98.4% recall and precision rate. A novel method for classifying breast cancer is presented in paper [130] and makes use of beta wavelet autoencoders (BWAE) and fully convolutional networks (FCNs). Known for its powerful image segmentation skills, FCN is used to identify key zones for modeling and extract meaningful information from mammography pictures.…”

“…When manually engineering features becomes a challenge, this may be quite helpful. To categorize nuclei and non-nuclei patches retrieved from BC histology, for instance, Jun Xu et al (2014) [147] used SSAE to get knowledge of important high-level traits for better input raw data representation. • Noise Reduction: Autoencoders can be used to remove noise from data.…”

“…Apart from the previously mentioned papers [139,140] that used SDAE networks, there are other ways to denoise the input data. [147], used a Beta Wavelet Autoencoder was used for feature extraction from mammography images, which inherently involves noise reduction. Also, a study by Manar Ahmed Hamza (2023) [148], mentioned that a Hybrid Denoising Autoencoder was specifically used for noise reduction in digital mammograms.…”