Artificial Intelligence harnessed to improve healthcare.
When it comes to devastating illness, there’s no technology too far fetched or out of reach. Thanks to unparalleled fundraising efforts to fuel research, the war against breast cancer has a new ally: machine learning. The AI capability to predict whether or not a biopsied lesion found in breast tissue will be malignant has reached a 97% success rate, leading to a 30% decrease in necessary surgery.
This is welcome news for patients and medical practitioners alike. In a test sample of more than 300 biopsies, the AI-based program was able to reach the aforementioned success rate based on a number of factors that had been supplied, including health of the patient, family history of breast cancer, and more.
Tools of the trade
While software will not replace standard medical protocols in diagnosing potentially fatal illnesses anytime soon, it does serve as another tool in the care team’s toolbox for proper treatment and positive patient outcomes.
“Because diagnostic tools are inexact, there is an understandable tendency for doctors to over-screen for breast cancer,” said the research paper’s co-author Regina Barzilay, Ph.D., the Delta Electronics Professor of Electrical Engineering and Computer Science at MIT. “When there’s this much uncertainty in data, machine learning is exactly the tool that we need to improve detection and prevent overtreatment.”
Some patients might balk at letting a computer determine whether or not their doctor takes more invasive steps, and it’s certainly alarming to think that health insurance companies or government-run medical facilities could decide you don’t need further care based on the opinion of a set of data. So far, though, this capability is cutting down on unnecessary surgery and patient anxiety, both of which are positive health outcomes.