Bridging Pixels and Practice: AI-Assisted Mammography as a Next-Generation Diagnostic Strategy for Breast Cancer Screening

TL;DR

AI-assisted mammography could improve breast cancer screening accuracy and efficiency but requires addressing legal, ethical, and systemic challenges for widespread adoption.

Contribution

This paper reviews the potential of AI-assisted mammography as a next-generation diagnostic strategy for breast cancer screening.

Findings

AI-assisted mammography improves cancer detection rates and diagnostic accuracy.

Challenges include legal, ethical issues and transferability across healthcare systems.

Stakeholders emphasize the need for human oversight and interdisciplinary collaboration.

Abstract

Breast cancer remains a leading cause of cancer-related mortality among women worldwide. Mammography is the cornerstone of population-based breast cancer screening, significantly improving prognosis and survival outcomes. However, limitations related to diagnostic accuracy, efficacy, and inequitable access persist. Recent advances in artificial intelligence (AI) have transformed AI-assisted mammography into a next-generation diagnostic strategy capable of enhancing screening performance within integrated diagnostic pathways. This narrative review examined whether AI-assisted mammography could serve as an effective next-generation diagnostic approach for breast cancer screening in global healthcare settings. Following PRISMA guidelines, studies were systematically screened using predefined eligibility criteria and quality appraisal with the CASP checklist, resulting in the inclusion of fourteen selected studies for analysis. Thematic synthesis identified three key domains: benefits, challenges, and stakeholder perspectives. AI-assisted mammography demonstrated improved cancer detection rates, enhanced diagnostic accuracy, reduced workload, and reduced recalls. However, challenges related to legal, ethical and social issues as well as transferability remain significant. Stakeholder perspectives emphasised the importance of human oversight, interdisciplinary collaboration, and workforce readiness. In conclusion, AI-assisted mammography offers significant promise as a next-generation diagnostic strategy, but successful translation into clinical practice requires substantial systemic, organisational, and practice-level changes across healthcare settings.

Figures