MACHINE LEARNING AS AN AID TO BREAST CANCER DETECTION AND PREVENTION: REALITY AND FUTURE PROSPECTS

Abstract

Breast cancer is one of the main global public health challenges, standing out as the neoplasm with the highest incidence among women. Early diagnosis is essential to increase survival rates; however, traditional methods face limitations such as a high false positive and false-negative. In this context, machine learning (ML) emerges as a promising tool for improving disease detection and prognosis. This literature review analysed studies published between 2019 and 2024 in the PubMed, Periódicos Capes, and SciELO databases, using specific descriptors related to breast cancer and ML. The findings indicate that ML techniques, such as deep neural networks and transfer learning, demonstrate high accuracy in classifying mammographic images, reducing the need for unnecessary biopsies, and improving cancer risk prediction.In addition, artificial intelligence has been applied to optimise screening exams, increasing diagnostic accuracy and minimizing variability between radiologists. However, challenges such as interpretability of the models and the need for robust clinical validation still remain. It is concluded that machine learning represents a significant advance in breast cancer diagnosis and prognosis, but continuous efforts are needed for its full integration into clinical practice.