Artificial Intelligence in Breast Cancer Diagnosis: A Review

The impact of human errors in imaging interpretation and the fact that decision support systems can improve the reliability and accuracy of radiology reporting have led to the more widespread use of these techniques. Developing decision support systems that assist radiologists in accurate diagnoses and improving the medical decision-making process has always been a challenge for the data analysis industry. This paper presents an in-depth review of a large number of intelligent approaches related to the support of breast cancer diagnosis taken in recent years. Specifically, the present review includes 230 corresponding approaches presented in the last 30 years in scientific journals in the field of artificial intelligence, as well as in medical journals. The search for the scientific reports included and presented in this paper was carried out in the databases of well-known scientific publishing houses using related keywords and phrases. The review briefly presents the main findings of each paper and classifies them according to their medical topic of specific interest (diagnosis, breast lesion classification, detection, abnormality classification, estimation of cancer risk, etc.). A statistical analysis is also provided, regarding the popularity of the approaches both, from the medical and AI viewpoint.