Deep Health - New AI model for metabolic analysis
Together with Lifespin and TUM, HSWT is developing an AI model that understands metabolic signals and opens up new avenues for personalised diagnostics.
The Weihenstephan-Triesdorf University of Applied Sciences (HSWT) is launching the „Deep Health“ research project together with the Technical University of Munich (TUM) and lifespin GmbH at BioPark Regensburg. The aim is to use artificial intelligence to decode the language of human metabolism. HSWT is contributing its experience in applied AI and data analysis. The project is being funded by the Bavarian Research Foundation with around 700,000 euros.
A better assessment of health
The three-year project combines the precision of nuclear magnetic resonance spectroscopy (NMR) with the latest machine learning approaches. The researchers are developing a model to interpret the „language“ of NMR spectra in a similar way to how language models - such as ChatGPT - understand natural language. The new technology should help to assess people's state of health more accurately, recognise changes in metabolism at an early stage and make better predictions of risks.
Prof Dr Andreas Krumpel and his team are working on making the models practical and usable for medical applications. Professor Krumpel emphasises: „The combination of state-of-the-art AI methods with NMR technology shows how interdisciplinary research enables innovation and at the same time strengthens Bavaria's position in the international life science sector.“ Once the project phase is complete, the plan is to transfer the results into practice and establish further research collaborations.
Further information on the start of the project can be found in this press release from BioPark Regenburg GmbH.
