Accuracy Analysis of Deep Learning Methods in Breast Cancer Classification: A Structured Review

Abstract

Breast cancer is diagnosed using histopathological imaging. This task is extremely time-consuming due to high image complexity and volume. However, it is important to facilitate the early detection of breast cancer for medical intervention. Deep learning (DL) has become popular in medical imaging solutions and has demonstrated various levels of performance in diagnosing cancerous images. Nonetheless, achieving high precision while minimizing overfitting remains a significant challenge for classification solutions. The handling of imbalanced data and incorrect labeling is a further concern. Additional methods, such as pre-processing, ensemble, and normalization techniques, have been established to enhance image characteristics. These methods could influence classification solutions and be used to overcome overfitting and data balancing issues. Hence, developing a more sophisticated DL variant could improve classification accuracy while reducing overfitting. Technological advancements in DL have fueled automated breast cancer diagnosis growth in recent years. This paper reviewed studies on the capability of DL to classify histopathological breast cancer images, as the objective of this study was to systematically review and analyze current research on the classification of histopathological images. Additionally, literature from the Scopus and Web of Science (WOS) indexes was reviewed. This study assessed recent approaches for histopathological breast cancer image classification in DL applications for papers published up until November 2022. The findings of this study suggest that DL methods, especially convolution neural networks and their hybrids, are the most cutting-edge approaches currently in use. To find a new technique, it is necessary first to survey the landscape of existing DL approaches and their hybrid methods to conduct comparisons and case studies.

Towards Agility in Breast Cancer Treatment Principles as Adopted from Agile Software Engineering

“…42 , 43 Recently, researchers have investigated the role of artificial intelligence and deep learning in detecting BC. 44 , 45 Innovation is becoming compulsory in the BCT field. 46 For all promising research findings, researchers propose their recommendations to professionals who work in the field, where these findings could lead to changes in the form improvements for their patients’ cases.…”

Recent advances of liquid biopsy: Interdisciplinary strategies toward clinical decision‐making

“…To ensure more reliable outputs from machine learning models, it's vital to: (i) Increase the number of participants; (ii) Establish training, validation and test groups to limit overfitting. 241 from a variety of bodily fluids. Thorough analysis allows us to extract and interpret the information carried by these biomarkers.…”

“…However, machine learning models require large, multidimensional datasets that encompass various patient information including gender, age, type and stage of cancer, proteome, metabolome, level of biomarkers, and information contained within biomarkers, to avoid overfitting‐induced impressive accuracy but potentially misleading results. To ensure more reliable outputs from machine learning models, it's vital to: (i) Increase the number of participants; (ii) Establish training, validation and test groups to limit overfitting 241 …”

Editorial on Special Issue “Medical Data Processing and Analysis”

“…The use of medical data to enhance the accuracy of breast cancer diagnosis is widely recognized, as breast cancer remains the most prevalent cancer among women globally. For example, M. Yusoff et al [ 12 ] systematically analyzed 25 papers published from 2018 to 2020 and evaluated their methods and performance metrics. The results showed that CNNs were the most used deep learning approach for breast cancer classification, followed by ensemble methods and recurrent neural networks (RNNs).…”

“…The study also highlighted limitations and challenges in the reviewed papers, such as limited sample sizes, lack of standardized datasets, and difficulty interpreting deep learning models’ decision-making process. In conclusion, the article provides a comprehensive overview of the current state of deep learning methods in breast cancer classification and identifies potential areas for future research [ 12 ].…”