AI for Health Data

Pipelines of clinical data from electronic health records to lab and imaging data and multi-OMICS platforms deliver –after data preparation and processing -significant medical information. To harness meaning from this heterogeneous and highly complex datasets we have created applied deep learning methods that researchers can analyze, deconstruct and train to make them suited to personalized medicine approaches.

Computer screen with machine learning estimating nutrients in food images (ARTORG Center, University of Bern)
Machine Learning, image segmentation and depth recognition can provide automated volume estimations. Here: An AI-based system to assess nutrient intake for hospitalised patients. (© ARTORG Center, University of Bern, and Inselspital, Bern University Hospital)