They are experimenting cancer diagnosis with Artificial Intelligence. The AI has detected more cases of breast cancer than a panel of expert doctors. The news comes from a Swedish study. The results can be read in the scientific journal The Lancet Oncology. Research has pointed out that the perfect combination is the collaboration between the AI and the expert doctor.
In this article
Dcancer diagnosis with Artificial Intelligence: Swedish research
The researchers put 80,000 under the magnifying glass mammography carried out in Sweden between April 2021 and July 2022. Half of the exams were reviewed by the AI, the other half by doctors with years of experience.
By comparing the results, the research group highlighted that AI discovered 20% more cases of breast cancer. The real advantage is that “the machines” do not suffer the stress of a heavy workload and save time.
Saving time is an important advantage
If radiologists read about 50 mammograms an hour, it would take a single radiologist four to six months less to read about 40,000 screening exams with the help of AI than it would take two radiologists alone. A little what happened when theAI has made it possible to develop a new antibiotic.
Dcancer diagnosis with Artificial Intelligence: radiologists happy with the results of this research
European guidelines require that each mammogram be examined by two radiologists. Among other things, this is a category of professionals that is not particularly numerous. Helping them read the data could be very important. So the results of this study were well received by the American Radiologists Association, which sees AI more as an aid to their work and not a threat. They let it be known that any tool that can facilitate their work is absolutely welcome.
Growing cases of breast cancer
The incidence of breast cancer is growing at 0.5 percent annually, according to the American Oncology Society. The data is very similar also in Italy. The survival rates for this cancer are good, thanks to the screening tests, which are able to make an early diagnosis.
Detecting breast cancer is extremely difficult, even after years of experience. A radiologist must locate a tumor that is white in the midst of a white background. AI may one day be able to help, even if the doctor remains central.