The Implementation of Artificial Intelligence in Breast Cancer Screening

Breast cancer screening is traditionally administered by radiologists through mammography. Most women should be receiving mammograms once they reach their 40s to detect any signs of cancer early, especially those with high risk. Recently, the development of AI in medicine has proven to be useful in several fields, one of which is breast cancer screening. This literature review looks into the current status of AI in this field and evaluates different models of different types that could contribute to diagnosing breast cancer. The methodology used was searching key terms on the database PubMed, and then compiling and organizing papers into a table that highlighted similar results, which were summarized and analyzed. It was found that AI has the potential to be an accessible option. Deep learning, in particular, has seen the most experimentation. The advantages of using this technology include using it as a complementary tool for double reading, improving the time efficiency for the radiologists without losing accuracy, and its inclusivity throughout all populations. However, there are still limitations because many proposals themselves discuss the work that needs to be done before they can be generalized. External validation studies haven’t yielded ideal results, exposing the complications of using AI. Ethics of using this technology are also important to consider because AI is replacing human tasks. Implementing AI in a prominent role in breast cancer screening should take more examination, but promising developments have already set the framework.