resumen

Mohamed El Gazzar*, Majed Salem Juaifer Bajuaifer, Amal Abdullah Ali Masmali, Amina Mohammed ali almomen, Norah Eid Hareer AlAnazi, Khulood Khalid Rashed Alrabyee, Israa Ahmed Salem Alhaffaf, Noor Abdullah Abdulrahim Altarooti, Ghadi Said Ali Alahmadi, Talal Mohammed Abu Jazilah, Fouz Mohammed Alshamrani, Nissren Tamam

Background: Breast cancer remains the most prevalent cancer among women worldwide, and accurate detection is essential for improving outcomes. Recent advancements in imaging, artificial intelligence (AI), and molecular diagnostics have significantly enhanced diagnostic capabilities, yet their clinical translation remains uneven.

Objective: To systematically review recent evidence on breast cancer detection technologies, focusing on imaging modalities, AI applications, and emerging biomarkers. Methods: A systematic review was conducted in accordance with PRISMA 2020 guidelines. Records were identified through database and manual searches (n = 427), with 395 remaining after deduplication. Following screening, 43 full-text articles were assessed, and 12 studies met inclusion criteria. Eligible studies included peer-reviewed research from 2016 to 2025 that evaluated imaging technologies, AI-based models, or molecular approaches for breast cancer detection.

Results: Evidence highlights the enhanced performance of tom synthesis, cone-beam CT, and automated ultrasound in complementing traditional mammography and handheld ultrasound. AI and deep learning models significantly improved lesion detection, classification, and risk assessment across modalities. Emerging approaches such as microwave imaging, Nano platform-based imaging, and extracellular vesicle biomarkers show promise for early detection. Despite these advances, challenges including dataset heterogeneity, cost, and limited clinical validation remain.

Conclusion: Advancements in imaging, AI, and molecular diagnostics are reshaping breast cancer detection. The future lies in multimodal, integrative approaches that combine anatomical, computational, and molecular data. Translation into clinical practice will require large-scale validation, standardization, and strategies to ensure accessibility across diverse healthcare settings.