Interview with Dr. Sindy Neumann, numares AG

I work for numares AG in the Biopark (Regensburg) and head the biostatistics department there.

First of all, I am responsible for the technical and personnel management of the department. But my tasks also include the strategic development in the area of ​​machine learning.

Numares develops diagnostic tests for clinical diagnostics and uses machine learning methods (e.g. supervised and unsupervised learning) to find patterns in the clinical data.

We always offer the opportunity for a job as working student and theses. And of course we are happy when interested students come to us. We would also like to expand our cooperation in the future as part of the AI ​​initiative of the city of Regensburg.

The advantage lies in the interdisciplinary exchange and the connection to basic research. Both are important factors to strengthen your own innovative ability. And finally, we would of course like to take the opportunity to inspire young talents for our company.

AI is already part of our everyday life and will become even more important and natural in the future. The earlier we introduce students to this topic, the better. In addition to conveying the wide range of possible applications, it is just as important to point out the challenges and limits.

We develop diagnostic tests for various clinical indications such as cardiovascular diseases, nephrology, oncology and neurology.

Our diagnostic tests are based on a specific interplay of metabolism-based biomarkers. We call this metabolite constellation. In order to discover this specific constellation for the respective question, we need methods from machine learning. We work a lot with algorithms from the field of supervised and unsupervised learning.

I studied bioinformatics and used various machine learning methods in my doctorate. Since then, I have tried to keep my knowledge of AI as up-to-date as possible through conferences and publications and to apply it accordingly in our daily work.

That depends on the exact method behind it. The fields of application can be very different and diverse. Personally, I have worked relatively little with deep learning so far and still find these topics very challenging if you want to understand the exact theory behind it.

The variety of possible uses and the resulting potential.

–

I would like the course offerings to be made available to many students and not to concentrate on individual sub-areas. My aim is not that everyone should become an AI expert, but everyone should understand the principles of the associated methods, as well as the associated challenges and risks. Perhaps one can also consider organizing lectures for interested citizens. To make AI competences widely available.

For me the project is still relatively new and I haven’t had any points of contact so far. So I’m curious to see how it evolves.

In my opinion, it is important to always look at both sides. I can understand the criticism of AI. Of course, AI offers many and sometimes new possibilities. At the same time, we also face new challenges. In diagnostics, for example, ethical framework conditions have to be met and the data quality plays a central role for the success of AI systems. In addition, one should always critically question what one wants to achieve with AI and whether its use makes sense in a specific case. AI hype or not.


Thank you for taking the time to answer these questions. We wish you a nice day!


With pleasure. I wish you that too!