More than 39 million mammograms are performed every year, according to data from the Food and Drug Administration. Today, researchers are training a computer to read mammograms as well as a doctor. An article published by npr.org highlighted Regina Barzilay’s technological journey.
After Barzilay was diagnosed with breast cancer, she had trouble understanding the lack of advanced technology the health care industry obtained. At the end of her treatment, she began to research a technological loophole; a way for computers to read mammograms.
She collaborated with Connie Lehman, a Harvard University radiologist who is the chief of breast imaging at Massachusetts General Hospital. According to the article, they developed an algorithm to determine breast density and when Lehman read a mammogram and assessed the density and then tested it on the computer, the assessments matched.
NPR points out that the biggest challenge isn’t the technology, but ethics. Researchers are concerned if the algorithm makes a mistake, who should take on the responsibility? The medical/legal aspect has to be solved before this technology can get started, but having the technology that has the ability to do this is a great start in medical innovations. After all, computers remember more than humans do!
According to the end research results, in 2017, Barzilay, Lehman, and colleagues proved that their algorithm could reduce biopsies by about 30% and they also developed a computer program that reviews information about a patient to predict potential future risk of breast cancer.
Barzilay explained that the model informs you of the likelihood of you developing cancer within two years, three years, and even ten years.
MEDcraze encourages you to ask questions and find out more technologies that could help your situation. Visit www.medcraze.com today and ask our experts about innovations such as this one!