AUTHOR=Hussain Sadam , Ali Mansoor , Naseem Usman , Nezhadmoghadam Fahimeh , Jatoi Munsif Ali , Gulliver T. Aaron , Tamez-Peña Jose Gerardo 

TITLE=Breast cancer risk prediction using machine learning: a systematic review

JOURNAL=Frontiers in Oncology

VOLUME=Volume 14 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2024.1343627

DOI=10.3389/fonc.2024.1343627

ISSN=2234-943X

ABSTRACT=Background: Breast cancer is the leading cause of cancer-associated fatalities among women globally. Conventional screening and risk prediction models primarily rely on demographic and patient clinical history to devise policies and estimate likelihood. However, recent advancements in artificial intelligence (AI) techniques, particularly deep learning (DL), have shown promise in developing personalized risk models. These models leverage individual patient information from medical imaging and associated reports. In this systematic review, we thoroughly investigate the existing literature on the application of DL in digital mammography, radiomics, genomics, and clinical information for breast cancer risk assessment. We critically analyze the studies and discuss their findings, highlighting the promising prospects of DL techniques in breast cancer risk prediction. Additionally, we explore ongoing research initiatives and potential future applications of AI-driven approaches to further improve breast cancer risk prediction, thereby facilitating more effective screening and personalized risk management strategies.Objective and methods: This paper presents a comprehensive overview of both imaging and non-imaging features used in breast cancer risk prediction via traditional and AI models.The features reviewed in this study include imaging, radiomics, genomics, and clinical features.Furthermore, this survey systematically presents DL methods developed for breast cancer risk prediction, aiming to be useful for both beginners and advanced-level researchers.Results: A total of 600 articles were identified, out of which 20 studies met the set criteria and were selected. Parallel benchmarking of DL models, along with natural language processing (NLP) applied to imaging and non-imaging features, could allow clinicians and researchers to gain greater awareness as they consider clinical deployment or development of new models. This