Department of Diabetes, Endocrinology, Nutritional Medicine and Metabolism

Inselspital Bern

The Department delivers comprehensive care, diagnosis and treatment for all forms of diabetes, endocrine and metabolic disorders, nutritional and obesity-related diseases. UDEM is committed to teaching and translational as well as clinical research, from hormone and metabolism studies to bioengineering and data science projects.

To the Department website

Director 

Prof. Christoph Stettler

Profile

  • Thirteen research groups in the areas of  glucose and metabolic regulation, diabetes technologies, obesity and nutritional medicine, neurocognitive aspects of endocrine and metabolic diseases, inborn errors of metabolism, sex- and gender-specific metabolic health, clinical endocrinology and diabetology, as well as artificial intelligence, data science, and decision support for metabolic health.
  • Undergraduate, graduate, and post-graduate education in Metabolic Science
  • Postgraduate education (CAS in Sex and Gender-specific Medicine, Certificate Course Clinical Nutrition)

External Partners

University of Bern, University of Basel, University of Zurich, University Hospital Zurich, ETH Zurich, ETH Lausanne (EPFL), CHUV (Lausanne University Hospital), Università della Svizzera italiana (USI), CSEM, University of St. Gallen, University of Cambridge (UK), University of Manchester (UK), Swansea University (UK), University of Cologne (DE), Technical University of Munich (DE), University of Tübingen (DE), FAU Erlangen-Nürnberg (DE), TU Dresden (DE), University of Lyon (FR), Erasmus University Rotterdam (NL), University of Padova (IT), University of Graz (AT), University of Girona (ES), University of Toronto (CA), McGill University (CA), Duke University (USA), Yale University (USA), University of California, Berkeley (USA), University of Virginia (USA), Industry Partners

Grants

  • SNSF project grants: Decipher the glp-1 estradiol crosstalk to tackle postmenopausal diabetes (Prof. L. Bally), Brain Aging in Patients with Phenylketonuria - a Longitudinal Study (Prof. R. Everts, Prof. Dr. R. Trepp), Porphyrin-based Aggregation-induced Emission Generating Molecules for Continuous Multi-Metabolite Monitoring (Prof. L. Witthauer), EMPASTONE – Randomized placebo-controlled trial to assess efficacies of EMPAgliflozin and personalised dietary counselling for kidney STONE prevention in patients with calcium kidney stones (Prof. L. Bally)
  • InnoSuisse Projects: Clinical AI Companion for Diabetes Optimisation using Digital Data from Continuous Glucose Monitoring. (Prof. L. Bally), Revolutionizing Swine Monitoring with Radar Technology and AI (Prof. L. Witthauer), Bern Medtech Collaboration Call Projects (Prof. C. Stettler)
  • SF Board Call Projects: Towards a FemTech Digital Twin approach to improve Women’s Health (Prof. L. Bally), Precision Metabolic Medicine Network by the Medical Faculty (Prof. J. Garcia), Swiss Precision Digital Therapeutics for the Prevention of Type-2 Diabetes - InnoSuisse Flagship (Prof. L. Bally), AI Nutritionist (Prof. L. Bally), FemTrack: A point of care platform for female hormone monitoring from finger prick (Prof. L. Bally)
  • Further funding sources: Bangerter Rhyner Foundation, Bernischer Hilfsbund, Boehringer Ingelheim ISS, Dexcom Inc ISS, DLF, Diabetes Center Berne DCB, Foundation Johanna Dürmüller-Boll, Foundation Rolf Gaillard, Fondation Pierre Mercier, Frieda Locher-Hofmann Foundation, Helmut Horten Foundation, Nestlé Health Science ISS, Novartis, NovoNordisk ISS, Nutricia ISS, Swiss Diabetes Foundation, Swiss Heart Foundation, UniBern Forschungsstiftung, VonTobel Foundation, Walter Fuchs Foundation, ATX Suisse, Burgergemeinde, CSEM, Maiores Stiftung, Novartis, Philhumana Stiftung, SAMW, Seerave Foundation, Siegenthaler Stiftung, QUMEA AG, Ypsomed

Highlights 2025

A real-time digital twin for type 1 diabetes using simulation-based inference

This study presents a real-time digital twin for people with type 1 diabetes, built on the machine learning paradigm of simulation-based inference. Trained purely on simulated glucose, insulin and meal data, the machine learning model rapidly personalises key physiological parameters from continuous glucose monitoring data. Compared with a Markov Chain Monte Carlo baseline, it yields more accurate parameter estimates, better calibrated uncertainty and improved prediction in counterfactual scenarios. This digital twin framework demonstrates an efficient and data-driven approach to personalised glucose dynamics modelling.

Hoang et al., Springer Nature. 2025

Listening to hypoglycemia – detection of a medical emergency from voice

Hypoglycemia is a dangerous diabetes-related emergency. In this study, voice data was collected using a smartphone during controlled hypoglycemia. Machine learning models showed high accuracy in detecting hypoglycemia, corroborating the potential of machine learning to infer acute health states through voice.

Lehmann et al., Diabetes Care. 2026

Launching MeHDI and ORN: a year of collaboration and growth

Led by UDEM and URI, the two collaborative initiatives Metabolism, Digital Health, Inflammation (MeHDI) and the Osteology Research Network (ORN) officially launched this year. Together, they aim to drive interdisciplinary research across four key areas: musculoskeletal health, metabolic disease, inflammation, and digital health. Major milestones included the joint MeHDI-ORN kick-off meeting in August, strong representation at University of Bern’s Nacht der Forschung in September, and the patient–public outreach event Erleben.Verstehen.Mitgestalten. In October, ORN introduced the DIMO-S grants to support interdisciplinary systematic reviews in the four focus areas, with results expected in 2026.

To the MIDHAS website of the University of Bern