New technology gets it right 99 times out of a 100
Researchers at Houston Methodist Research Institute have designed an artificial intelligence (AI) robot that is better than doctors at detecting breast cancer. Researcher Dr. Stephen Wong, who is behind the experiment, explained that the software is designed to continually get smarter, and that it can go through millions of records in a very short time. Using the method, the artificial intelligence system gets smarter by comparing it’s guesses to a huge database of both research and previous cancer samples to better determine whether women’s mammograms reveal either cancer or cancer risk.
30 times faster than human doctors
By analyzing and cross referencing doctors’ medical notes, previous research, and the patient’s personal medical history, the software can now diagnose breast cancer with 99 percent accuracy. That is 20 times more accurate than human doctors can handle. In the latest experiment, researchers tested the software on 500 breast cancer patients, and in a few hours, the AI system identified all women with specific subsets of cancer – 30 times faster than the doctors normally do it.
Every second woman gets scared for no reason
Mammography is a widely used method to detect breast cancer in many countries. The method has its limitations, and in the US every second woman who is examined gets the message that they might have cancer after aexamination. The new technology may prevent a large number of women being called to unnecessary follow-up biopsies when the doctor’s suspicion of cancer is unfounded.
20 percent of biopsies are unnecessary
12.1 million mammograms are performed every year in the United States alone, according to statistics from the Centers for Disease Control and Prevention (CDC). Statistics from American Cancer Society (ACS) show that fifty percent of all mammograms show false positive results. More than 1.6 million biopsies are performed per year nationwide and about 20 percent of these turn out to be unnecessary due to false-positive diagnoses.
Read more about the study here
