AI Center Inaugural Datathon

The new Technion-Rambam Center for Artificial Intelligence in Healthcare (CAIH) – a joint initiative of Rambam Health Care Campus and the Technion – Israel Institute of Technology – organized a Datathon – an information-based competition – as its inaugural event. Eight teams and 50 participants worked on four challenges proposed by Rambam physicians. The event took place in the Technion Faculty of Biomedical Engineering.

“The clinicians bring their side, which is formulating the problem and the clinical knowledge, while the students and researchers from the Technion bring their side, which is how to build algorithms and analyze big data. Together we’ll find solutions,” said Assistant Professor Danny Eytan.

The teams were asked to develop a model based on computational learning to solve clinical problems encountered by hospital physicians, using real anonymized data from the hospital. The groups included students and graduates from eight faculties at the Technion, medical personnel, and industry personnel.

The winning team of students. From left to right: Prof. Leo Anthony Celi, the winning students (in white shirts), the facilitator Eytan Katz, Dr. Jonathan Sobel, Dr. Joachim Behar, Dr. Ronit Almog, Asst. Prof. Danny Eytan Photo credit: Yossi Weiner

The list of winners is as follows:

The winners were Team Stem Cells, who presented a model for early prediction of blood infection in bone marrow transplant patients. Team members: Omer Shubi, Tom Yuviler, Oren Ploznik, Yoav Danieli, Yotam Martin, Nitzan Dahan, and Shoval Zandberg. The team was mentored by Eytan Kats, Israel Henig, and Asaf Miller.

Second place went to Team Birth, who presented a model for the personalized prediction of birth weight (as an indicator of future defects) based on clinical parameters and previous births. The team was comprised of Anastasiya Kuznetsova, Alon Hacohen, Noam Keidar, Rotem Shapira, Galya Segal, and Shiri Fistel. The team was mentored by Marie-Laure Charpignon, Pierre Aublin, and Ron Beloosesky.

In third place was Team COVID-19, who presented a model for early prediction for Coronavirus patients regarding the chances of recovery, and the number of days artificial ventilation would be required. The team members were Yotam Granov, Michal Jacob, Hadar Guthmann, Alon Tsaizel, Ofek Avraham, Gal Binary, and Hagay Michaeli. The team was mentored by Einat Borohovich and Danny Eytan.

Technion and the Rambam IT and Epidemiology Department worked hand in hand to set up the computational cloud infrastructure and collate original datasets from several units across the hospital. The Datathon was a joint effort between the Technion, Rambam Health Care Campus, and the Massachusetts Institute of Technology (MIT). It was organized by staff from all three institutions: Assistant Professor Joachim Behar (Technion), Assistant Professor Danny Eytan (Technion and Rambam), Dr. Ronit Almog (Rambam), Professor Leo Anthony Celi (MIT), and Dr. Jonathan Sobel (Technion). The event was also supported by several academic and industrial partners including Roche, GE Healthcare, and Technion Human Health Initiative (THHI).

Video from the datathon: