TwoViewDensityNet: Two-View Mammographic Breast Density Classification Based on Deep Convolutional Neural Network

Abstract

Dense breast tissue is a significant factor that increases the risk of breast cancer. Current mammographic density classification approaches are unable to provide enough classification accuracy. However, it remains a difficult problem to classify breast density. This paper proposes TwoViewDensityNet, an end-to-end deep learning-based method for mammographic breast density classification. The craniocaudal (CC) and mediolateral oblique (MLO) views of screening mammography provide two different views of each breast. As the two views are complementary, and dual-view-based methods have proven efficient, we use two views for breast classification. The loss function plays a key role in training a deep model; we employ the focal loss function because it focuses on learning hard cases. The method was thoroughly evaluated on two public datasets using 5-fold cross-validation, and it achieved an overall performance (F-score of 98.63%, AUC of 99.51%, accuracy of 95.83%) on DDSM and (F-score of 97.14%, AUC of 97.44%, accuracy of 96%) on the INbreast. The comparison shows that the TwoViewDensityNet outperforms the state-of-the-art methods for classifying breast density into BI-RADS class. It aids healthcare providers in providing patients with more accurate information and will help improve the diagnostic accuracy and reliability of mammographic breast density evaluation in clinical care.

A novel fusion framework of deep bottleneck residual convolutional neural network for breast cancer classification from mammogram images

“…In this table, the authors in (30) obtained an accuracy of 83.19%, later improved by authors in (42) at 95.6%. The other listed methods in this table, such as (43)(44)(45), obtained accuracies of 95.1%, 93.0%, and 96.0%, respectively. The proposed method obtained an accuracy of 96.5%, which is better than the recently presented methods.…”

A Systematic Literature Review on Mammography: Deep Learning in Redefining Breast Cancer Diagnosis for the Asian Perspective

“…This classification can be multiclass or binary. (Chugh et al, 2022;Razali et al, 2023a) Explained in Table 10 ( Razali et al, 2023a) Explained in Table 10 Morphological, Shape (Chugh et al, 2022;Razali et al, 2023a) Explained in (Li et al, 2020), (Saffari et al, 2020;Nithya and Santhi, 2021;Ciritsis et al, 2019;Mohamed et al, 2018a,b;Chugh et al, 2022;Diniz et al, 2018;Busaleh et al, 2022;Pawar et al, 2022) In weakly supervised learning, ML models are trained on noisy or partially labeled data, with few or no detailed annotations available for each instance. It addresses the difficulties of learning from noisy or incomplete labels, allowing models to predict even in the presence of limited labeling data.…”

“…Convolutional layers are typically used to capture spatial features, pooling layers to downsample, and fully connected layers to classify data. Latest few studies have used dual view CNN (Busaleh et al, 2022;Pawar et al, 2022;Li et al, 2020) These networks are designed for dealing with graphstructured data, with entities represented as nodes and relationships as edges. GCNs efficiently detect dependencies among interconnected nodes.…”

“…Typically, they ignore the intrinsic relationship between two-view images taken from the same patient and treat them as separate entities. Recent advances in lesion classification (Li et al, 2020;AlGhamdi and Abdel-Mottaleb, 2021;Wu et al, 2019) , density classification (Busaleh et al, 2022;Pawar et al, 2022) have solved this problem. These initiatives take a more accurate and comprehensive approach by processing the two views in a parallel model and making decisions using combined data from both.…”

Mammographic Breast Composition Classification Using Swin Transformer Network

“…In recent times, numerous groups have proposed solutions for breast density classification based on deep learning, yielding promising results. (9)(10)(11)(12)(13) Diverging from previous studies, we adopt an exploratory approach using the Swin Transformer, (14,15) a foundation model for image classification, to classify the four-category breast density. The Swin Transformer has also demonstrated exceptional performance in the field of image classification.…”