Regina Barzilay teaches computers how to learn. A professor at the Massachusetts Institute of Technology, her work focused on natural language processing – training computers to understand human speech – until a breast cancer diagnosis three years ago.
"Going through it, I realized that today we have more sophisticated technology to select your shoes on Amazon than to adjust treatments for cancer patients," Barzilay said in an interview at her office in Cambridge. "I really wanted to make sure that the expertise we have would be used for helping people."
Barzilay's group, in collaboration with Massachusetts General Hospital, is now applying their expertise in artificial intelligence and machine learning to improve cancer diagnosis and treatment. They're asking questions like whether computers can detect signs of breast cancer in mammograms earlier than humans are currently capable of, and whether machine learning can enable doctors to use all the huge quantities of data available on patients to make more personalized treatment decisions.
It's a field some say is on the cusp of changing medicine.
"The potential is perhaps the biggest in any type of technology we've ever had in the field of medicine," said Dr. Eric Topol, director of the Scripps Translational Science Institute. "Computing capability can transcend what a human being could ever do in their lifetime."
Investment is pouring in, from tech giants like IBM's Watson, Alphabet and Philips, to pharmaceutical companies and swiftly proliferating startups. The market for artificial intelligence in health care and the life sciences is projected to grow by 40 percent a year, to $6.6 billion in 2021, according to estimates from Frost & Sullivan.
Some of the earliest applications are expected to be in diagnosing disease. For Barzilay, the ability of computers to scour images holds the potential of earlier detection. She had been getting mammograms for more than two years before she was diagnosed at age 43.