Wir überwinden Grenzen

The collaborative project “AI-COMPASS” is investigating how digital health data can be used to detect cardiovascular complications early on following treatment for gastroesophageal cancer

A collaborative project involving the Medical Center – University of Freiburg is investigating how smartwatch data and artificial intelligence can support the follow-up care of patients after gastric-esophageal cancer. With the help of artificial intelligence (AI), health data from smartwatches will be linked to routine clinical data and analyzed. This will enable physicians to better assess health risks. The “AI-Compass” project is funded as part of the “Interactive Technologies for Health and Quality of Life” funding program of the Federal Ministry of Research, Technology, and Space and the National Decade Against Cancer. 

“Many patients do not simply return to good health after cancer treatment. They need follow-up care that also identifies long-term effects and supports them in their daily lives,” says Prof. Dr. Michael Quante, head of the Department of Gastrointestinal Oncology at the Department of Medicine II at the Medical Center – University of Freiburg. He is the medical director of the project. “With this project, we aim to link clinical data with continuously collected health data in a way that allows physicians to identify risks earlier and better support patients. The AI is not intended to make decisions, but rather to provide understandable and transparent assistance.”

Humans and AI Work Together Transparently

Follow-up care after gastroesophageal cancer has so far focused primarily on detecting a recurrence early. However, many patients also experience long-term symptoms after treatment. These can include cardiovascular problems. Such issues are not always systematically recorded during routine follow-up care. The collaborative project is therefore investigating whether health data from smartwatches and smartphones can usefully supplement clinical follow-up care.

The smartwatch is designed to collect data on heart rate, physical activity, and sleep. This sensor data is combined with routine clinical data, diagnoses, and information on treatment. This creates digital patient models intended to help better identify individual risk patterns using AI. The AI operates in an explainable and transparent manner so that medical teams can understand the identified risks.

Joint Innovation Initiative

AI-COMPASS brings together experts from the fields of healthcare, data science, artificial intelligence, and patient advocacy. The consortium includes Hahn-Schickard, BiSigma GmbH, three specialized centers at the Medical Center – University of Freiburg (Center for Gastrointestinal Tumors, Heart Center, and Data Integration Center), as well as Digestive Cancers Europe (DiCE). Mario Cypko of the Hahn-Schickard Institute serves as the overall project lead for KI-COMPASS. The project is funded by the Federal Ministry of Research, Technology, and Space under grant number 16SV9610.