AI in Breast Cancer Classification and Treatment
Breast cancer is the most common cancer found in women, and despite many advancements and adjustments in therapy, it still poses a major concern for cancer-related mortality, contributing to over 500,000 deaths annually worldwide. Breast cancer screening programs using mammography are effective in reducing breast cancer-related mortality.
The problem lies in the high labor intensity in screening programs today as there are a large number of women that have to be screened. This problem, on top of growing concerns in relation to the scarcity of breast cancer screening radiologists in many countries has led to researchers believing that more efficient and less labour intensive breast cancer screening methods need to be researched and investigated.
In a study conducted recently, researchers are comparing, the cancer detection performance of a commercially available AI system to 101 radiologists who scored nine different cohorts of mammography examinations from four different manufacturers. Each set of data consisted of mammography exams obtained with systems from four different vendors, multiple radiologists' assessments per exam, generating a total of 2,652 exams and interpretations by 101 radiologists.
The performance of the artificial intelligence system was shown to be statistically superior to that of the average of the 101 radiologists. The AI system achieved a cancer detection accuracy comparable to an above-average breast radiologist.
One of the research authors went on to say that "before we could decide what is the best way for AI systems to be introduced in the realm of breast cancer screening with mammography, we wanted to know how good can these systems really be.”
The future looks promising for cancer treatments as AI is being integrated at a faster rate in different aspects of medicine.