Machine learning models - interview with DMEA young talent award winner

Numerous students once again applied for the DMEA Young Talent Award this year - Julian Hugo from the Hasso Plattner Institute Potsdam was able to beat off the competition in the Master's thesis category. He wrote his master's thesis in collaboration with the Hasso Plattner Institute and the Mount Sinai Health System (MSHS) in New York City. His goal: to use machine learning (ML) to create models that can predict the risk of Crohn's disease, a chronic bowel disease.

Crohn's disease is a complex disease that can severely impact the lives of those affected. Julian accessed structured data from MSHS electronic medical records to develop his models. As a first step, he designed rules to identify patients who already suffer from Crohn's disease. He then trained the ML models to predict the risk of the disease in other patients. He also focussed on identifying the relevant clinical variables that make these predictions particularly reliable.

From thesis to practice

After winning the young talent award, Julian seamlessly continued his work in the field of clinical decision support. ‘I am very pleased that I can actually continue to pursue the topic of Clinical Decision Support (CDS) at Tiplu GmbH. At Tiplu, we have great expertise in preparing data from hospital information systems in a syntactically and semantically interoperable way so that it can be used for the development of risk prediction models, for example. We are focussing in particular on calculating risks that can arise during a hospital stay, such as blood poisoning (sepsis) or a state of confusion (delirium).’ Julian's aim is to help doctors avoid these risks. The software is currently in the process of being certified as a medical device.

First DMEA then DMEA Young Talent Award

Students can visit the DMEA free of charge - Julian came to the DMEA in Berlin for the first time a few years ago as a student and got to know and love the digital health sector: ‘Even before and during my medical studies, I was interested in IT topics and built my own small software projects. Towards the end of my studies, it became increasingly clear to me how great the potential of digitalisation is in healthcare. I then fulfilled my dream of learning the technical basics of software development and machine learning with a Master's in Digital Health at the Hasso Plattner Institute in Potsdam. At the same time, I focused on the interoperability of healthcare data and data analyses. The Master's degree now enables me to work directly at the interface between software development and users.’

Julian found out about the young talent award via the DMEA newsletter. It quickly became clear to him that he wanted to apply for it with his thesis and he recommends that other students do the same: ‘All graduates have already overcome the biggest hurdle to applying: developing and writing their theses. I can only advise everyone to apply for the DMEA Young Talent Award: Getting to know the other graduates, the jury members and the digital health universe in general at the DMEA provides exciting insights and great opportunities for further development!’

By the way: The call for entries for the DMEA Young Talent Award 2025 starts in October. All information: https://www.dmea.de/en/newcomers/newcomer-award