Meet our new group leader from the unit Information at Forschungszentrum Jülich and his research on developing artificial intelligence methods for large-scale microstructural mapping of the human brain.
With a background in scientific programming and technomathematics, Schiffer has worked as a doctoral and postdoctoral researcher in the Institute of Neuroscience and Medicine (INM-1) at the Forschungszentrum Jülich. He shares his passion for applying novel AI methods to highly relevant neuroscientific research questions and implementing these methods on the fastest supercomputing systems in the world.
- What were your previous scientific stations?
2013-2016: Bachelor of Science “Scientific Programming”, University of Applied Sciences Aachen, Germany.
2016-2018: Master of Science “Technomathematik”, University of Applied Sciences Aachen, Germany.
2016-2018: Student position, Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Germany.
2018-2022: Doctoral student, Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Germany.
2022: Research stay (3 month), Montreal Neurological Institute (MNI), Canada.
Since 2022: Postdoctoral researcher, Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Germany.
Since 2023: Team leader “Large-scale AI for Brain Mapping”, Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Germany.
Since when and in which function have you been at Helmholtz AI?
I am a member of the Helmholtz AI research group in Jülich since its inception in 2019. From 2019 to 2022, I worked as a doctoral student, with the goal to develop deep learning algorithms for automatic classification of cytoarchitectonic brain areas from high-resolution microscopic images of histological human brain sections. Since 2022, I continue my research on this topic as a postdoctoral researcher in the Jülich Helmholtz AI research group. Since 2023, I lead the Young Investigator Group “Large-scale AI for brain mapping” in Jülich. Our research focuses on the development of artificial intelligence methods for large-scale microstructural mapping of the human brain.
What (research) goals are you pursuing there?
In my research, I aim to develop methods for data-driven holistic characterization of human brain architecture at microscopic scale, with the ultimate goal to deepen our understanding of the brain’s microstructural organizational principles. This goal encompasses the development of artificial intelligence methods to analyze multi-modal microscopic image data, as well as the design and implementation of distributed workflows for processing petabyte-scale microscopic image datasets.
What inspires you to come to Helmholtz AI? What is typical of Helmholtz AI for you?
I am in the special position to be a part of the Helmholtz AI research group in Jülich since it’s inception in 2019. In my view, one unique aspect of the Helmholtz AI network is that it brings together researchers from a wide range of different scientific domains, which encourages them to identify common challenges and share solutions. I can’t count how many times I had a “Oh, we actually have exactly the same problem!” moment after discussing with colleagues from entirely different scientific domains.
What fascinates you about your research?
My research allows me to work at the intersection of three extremely interesting fields: I apply novel artificial intelligence methods to highly relevant neuroscientific research questions and implement these methods on the fastest supercomputing systems in the world. Dealing with this combination of methodological, scientific, and technological aspects makes the research challenging, but at the same time very fascinating and rewarding.
What would you like to achieve in your scientific field?
I aim to provide neuroscientists with methods that enable them to analyze extremely large multi-modal microscopic imaging datasets in a holistic and data-driven way. I hope that this work will help to deepen our understanding of the brain’s structural and functional organization, and thereby shed light on the mechanisms that enable us to perform complex cognitive tasks.
What are the biggest challenges and why is it still worth it every day?
As mentioned before, I work at the intersection of artificial intelligence, neuroscience, and high-performance computing. Each of these fields comes with its own challenges: Keeping up with the rapid developments in AI, understanding neuroscientific research questions, and dealing with complex large-scale computing systems are challenging on their own. In combination, these challenges amplify each other even more, for example when novel AI methods need to be adapted for specific neuroscientific research questions and re-designed to run efficiently on terabyte- or petabyte-scale datasets. However, finding novel solutions for these unique challenges and observing how the obtained methods and result bring us exciting new insights makes it worth every day.
Was there a formative experience in your scientific career that left mark on you?
I cannot name a specific event, but my first month working in science was really eye opening for me. My first career steps were as a software developer in a company. I learned a lot there – many of the skills I still apply every day – but I noticed that working in industry might not be the right thing for me. When I started my first student position in the research group in Jülich, the first few weeks were overwhelming, because almost every day confronted me with something new to learn. Today I think that it is this wide range of different topics, tasks, methods, technologies, colleagues, and challenges that makes my work very fascinating and rewarding.
In your view, what characterizes the life of a scientist?
Curiosity, dedication, creativity, flexibility, and coffee.
What do you draw strength from next to your work? What hobbies do you have?
I would be lying if I would say that I try to stay away from computers to have a counterbalance for my work. I enjoy video games, programming, tinkering with hardware, and working with my 3D printer. Apart from that, I like reading books, hiking, and playing Dungeons & Dragons with my friends.
If you could meet three personalities of your choice for dinner - who would they be?
Linus Torvalds (developer of Linux), Guido van Rossum (developer of the python programming language), and Matthew Mercer (game master for the D&D show Critical Role). It might turn into a deep technical discussion, or an interesting D&D session. Both are fine for me.
What advice would you give your 20-year-old self today?
Don’t postpone the things you want to do until there is “more time”. There will always be that one important exam, thesis, paper, presentation, or project proposal that has to be finished. Try to find time now.
Tell a secret about yourself!
My go-to music to listen to while doing focused work: Disney soundtracks. Try it.