The health group at Microsoft Israel Research and Development has developed a project called New Hope, based on AI, which connects patients, doctors and health organizations to clinical trials that take place around the world, in order to overcome the difficulty of finding suitable candidates. At any given time, there are about 50,000 clinical trials around the world, which are necessary to create drugs and treatments for various medical conditions, including the corona virus, half of which fail to recruit potential participants.
“Clinical trials are not an alternative to traditional therapies, but they can certainly give hope to patients, who sometimes have difficulty understanding the criteria of these trials, and understanding whether they are appropriate,” explains group leader Hadas Bittern. “We want to prevent a situation where important experiments, such as those conducted today as part of corona research, fall because no participants were found. The technology we have developed helps advance the medical world, and we have found many thousands around the world who have agreed to volunteer for the various experiments.”
Bittern has worked for years at the Microsoft Development Center in Israel. “I am a graduate of Unit 8200 but I have always been interested in medicine,” she says. “I was even thinking of starting to study medicine, but at that point it did not fit, because I was a mother of small children. So I realized that my potential was in the development of health technologies, because I have always had an aspiration to help develop this field.” On the work areas of the group, which she set up five years ago, she says: “We combine artificial intelligence with the field of health, a not simple field. These days we are working on a lot of significant developments that are still impossible to tell.”
One of the developments the group worked on was called Text Analytics, in which the team led by it applied language comprehension models to all kinds of medical texts, descriptions of clinical trials, summarizing patient-physician encounters, medical articles and more.
“Technology enables healthcare organizations, healthcare technology organizations, researchers and more to extract information and insights that are ‘hidden’ in medical texts and thereby streamline healthcare systems and patient care,” she notes. “This is an analysis that understands elements in the text, is it a symptom, diagnosis, disease, cure? Is it mentioned in a positive or negative sense? It also connects the elements it has identified to medical coding systems, used by the whole world and Israel, and identifies connections between those elements.”
Bitter was invited to the White House in 2019, with two of the health group’s data scientists, Danny Carmon and Shachar Admati, to present its developments to senior US administration officials. “I was invited there along with some researchers from my team, as part of a federal government project aimed at developing projects and improving various health systems – and it has certainly promoted and helped a lot,” she says. “I am very proud of my group, and I know that one of the special things about Israelis is the combination of innovation and initiative.”
At the end of the conversation, she adds that “we are making a significant change, especially in the days of the Corona. Whenever the managers in the United States wanted something to happen quickly, they turned to the ‘Israeli commando force.’ For example, within two days we have developed a bot that helps check the suitability of plasma donations from Corona recoverers. ”