Welcome
Hi, I'm Daniel. Researcher and lecturer at Technische Hochschule Mittelhessen, Transfer Officer at hessian.AI, and freelance software developer.
At hessian.AI, I work at the interface of academia and industry — consulting, coordinating research projects, and writing proposals. I also lecture at THM.
My Ph.D. at Goethe Universität Frankfurt is on ontology-based transfer learning — using ontologies and knowledge graphs to classify medical texts, both clinical and colloquial.
Research Projects
- 2025 (ongoing) InNa – Intelligente Nabelklemme
Non-invasive wireless monitoring of neonatal vital signs via smart umbilical clamp with AI-based edge computing
Funded by Hessisches Ministerium für Digitalisierung und Innovation (Distr@l) - 2026 (ongoing) RareNavigator
AI-based matchmaking and navigation tool for stakeholders of rare diseases
Funded by BMWK (IGP) - 2026 (ongoing) FörderKompass
AI-based matching and application support for SMEs and impact companies
Funded by BMWK (IGP) - 2019 – 2023 Transfer Learning for Medical Diagnosis (TLDia)
LOEWE3 project
Publications
Transfer-Lernen für die Klassifikation medizinischer Texte
Published in Sammelband (Book Chapter), 2025
A book chapter on transfer learning methods for the classification of medical texts, building on ontology-based approaches to enable GDPR-compliant model reuse across institutions.
Recommended citation: Bruneß, D. et al. (2025). Transfer-Lernen für die Klassifikation medizinischer Texte.
A Hybrid AI-Based Method for ICD Classification of Medical Documents
Published in Healthcare Transformation with Informatics and Artificial Intelligence (IOS Press), 2023
A transfer learning method that uses ontologies to normalize the feature space of text classifiers to create a controlled vocabulary that ensures that the trained models do not contain personal data, and can be widely reused without violating the GDPR.
Recommended citation: Bruneß, D., Bay, M., Schulze, C., Guckert, M., & Minor, M. (2023). A Hybrid AI-Based Method for ICD Classification of Medical Documents. Healthcare Transformation with Informatics and Artificial Intelligence, 305, 1-4.. https://doi.org/10.3233/SHTI230408
An Ontology-based transfer learning method improving classification of medical documents.
Published in 2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA), 2022
A transfer learning method which uses ontologies to normalise the feature space of text classifiers to guarantee that the trained models do not contain any person related data and can therefore be widely reused without raising General Data Protection Regulation issues.
Recommended citation: Bruneß, D., Bay, M., Schulze, C., Guckert, M., & Minor, M. (2022). An Ontology-based transfer learning method improving classification of medical documents. 2022 21st IEEE International Conference on Machine Learning and Applications (ICMLA), 407-412. https://doi.org/10.1109/ICMLA55696.2022.00065
Term Extraction from Medical Documents Using Word Embeddings
Published in 2020 6th IEEE Congress on Information Science and Technology (CiSt), 2021
A new method for the extraction of discipline-specific terms from medical documents using word embeddings in a neighborhood context based method which is called Snowball because of its layerwise way of working.
Recommended citation: Bay, M., Bruneß, D., Herold, M., Schulze, C., Guckert, M., & Minor, M. (2020). Term Extraction from Medical Documents Using Word Embeddings. 2020 6th IEEE Congress on Information Science and Technology (CiSt), 328-333. https://doi.org/10.1109/CiSt49399.2021.9357263
Interacting Spider Webs
Published in The Art of Theoretical Biology, 2020
The bacterial pathogen Salmonella Typhimurium provokes gastroenteritis and typhoid fever. Salmonella become multidrug resistant.
Recommended citation: Rieser, J., Bruneß, D., Ackermann, J., Koch, I. (2020). Interacting Spider Webs. In: Matthäus, F., Matthäus, S., Harris, S., Hillen, T. (eds) The Art of Theoretical Biology. Springer https://doi.org/10.1007/978-3-030-33471-0_68
The new protein topology graph library web server
Published in Bioinformatics, 2016
A new, extended version of the Protein Topology Graph Library web server is presented, featuring additional information on ligand binding to secondary structure elements, increased usability and an application programming interface (API) to retrieve data, allowing for an automated analysis of protein topology.
Recommended citation: Schäfer, T., Scheck, A., Bruneß, D., May, P., & Koch, I. (2016). The new protein topology graph library web server. Bioinformatics, 32 3, 474-6 . https://doi.org/10.1093/bioinformatics%2Fbtv574
Teaching
Big Data & Data Wrangling
Seminar, Technische Hochschule Mittelhessen, StudiumPlus, 2023
Lecture in the B.Sc. Softwaretechnik (Data Science) program at THM StudiumPlus, 6 SWS / 6 CrP. Running since SoSe 2023, exam-based (90 min written).
Applied Natural Language Processing
Seminar, Technische Hochschule Mittelhessen, MND, 2021
Block seminar at THM (4 SWS), co-taught with Prof. Dr. Michael Guckert since WiSe 2021.
Mobile Technologies
Seminar, Technische Hochschule Mittelhessen, StudiumPlus, 2020
(dt. Mobile Technologien)
Algorithms and models in bioinformatics
Practical Course, Goethe University, Molecular Bioinformatics, 2014
(dt. Algorithmen und Modelle der Bioinformatik)
Programming 1
Tutorial, Goethe University, Computer Science, 2013
(dt. Programmierung 1)
Talks
A Hybrid AI-based Method for ICD Classification of Medical Documents
Conference Talk, ICIMTH '23 — 21st International Conference on Informatics, Management, and Technology in Healthcare, Athens, Greece
Short talk presenting our ontology-based transfer learning method for classifying medical documents with ICD codes. The core idea: instead of transferring statistical model weights — which risks leaking patient data under the GDPR — we normalize the classifier’s feature space through medical ontologies, making trained models reusable across institutions without additional training data.