Exciting insights and inspiring encounters at the Grob in-house exhibition in Mindelheim! This week we had the great opportunity to dive deep into the world of manufacturing and experience the latest technologies and processes up close.

Particularly impressive was our conversation with Emil Nigl, who presented Grob's advanced solutions for data acquisition in the industry. These technologies open up new possibilities for more precise and efficient manufacturing processes.

Another highlight was our exchange with Henning Strauß and Tim Johannsen from Kiel University of Applied Sciences. Together, we discussed innovative approaches to data collection and use in the production environment. The perspectives on digital twins and how they can revolutionize efficiency and precision in production were particularly fascinating. We also discussed the use of this data for artificial intelligence to further optimize our processes.

This experience not only broadened our understanding of advanced manufacturing technologies, but also provided valuable impulses for future projects and collaborations. We are excited to see what the future holds and look forward to integrating these insights into our work!

The SMDS team and colleagues from the KI-Produktionsnetzwerk have made significant progress in the utilization of machine data by recently completing two specialized trainings on the Orchestra software by Soffico. In the first training, which focused on the Orchestra Designer, participants were equipped with the necessary skills to effectively design data transmission scenarios. The Orchestra Designer, a key tool of the software, forms the basis for collecting and storing data, which will serve as the foundation for further data analyses in the future. The subsequent operations and monitoring training deepened the knowledge about managing and monitoring these scenarios through the Orchestra Monitor, another essential tool crucial for ensuring accurate and uninterrupted data collection. This combination of theoretical understanding and practical application with these tools paves the way for improved data integration and utilization, driving our production processes forward through technological innovation.

Our research assistant Tobias Foth delightfully participated in this year’s ACP Summer School on the topic ML4CP - Machine Learning for Constraint Programming.

In addition to solving (... and then enjoying) classic Belgian problems such as a satisfactory choice of sauces for the shared portion of fries the focus was on professional insights into current research  - enriched with lab session to try on some topics on our own and discussions with colleagues and lecturers in a relaxed atmosphere. After a poster session in the middle of the week in which participants could discuss their own research, there was even the opportunity to try out their own ideas in a small group and with professional support during a hackathon.

Despite a packed technical and exciting week, daily activities were also planned to get to know the city, the Belgian (food and drink) culture as well as the colleagues from various fields and countries - which was felt to be a great added value. In the end, this really well organized summer school was a wonderful and recommendable experience.

At the "International Conference on Software Technologies" (ICSOFT), Miriam Elia presented our paper titled "A Methodology based on Quality Gates for Certifiable AI in Medicine: Towards a Reliable Application of Metrics in Machine Learning". We introduce our methodology for certifiable AI in medicine and provide an exemplary approach of its implementation for binary classification metrics that measure the performance of machine learning models. We are pleased to share that our approach was honored with the Best Position Award.

Our paper, "Towards Automated COVID-19 Presence and Severity Classification," was presented at the Medical Informatics Europe (MIE) Conference. The paper introduces an approach based on the open-source framework Aucmedi, demonstrating how to classify the severity of COVID-19 based on lung CT scans. One potential use case for this approach could be in the planning of ICU bed allocation.

This year, Noël Hagemann and Florian Barthel presented the capacities of the autonomous vehicle developed at the chair in a one-hour demonstration at the Schülerinfotag. The so-called environmental model was a particular focus which depicts the car's view of its environment. The environmental model is one of the central components that enable the car to act autonomously.

We are very pleased to welcome Mr. Adrian Pfleiderer as a new scientific employee at the Chair of Software Methodology for Distributed Systems. He joined our team today and will be working on topics such as machine learning in production networks.

At the International Symposium in Biomedical Imaging (ISBI), Miriam Elia presented our abstract "Precision Medicine for Achalasia Diagnosis: A Multi-modal and Interdisciplinary Approach for Training Data Generation". There, an interdisciplinary approach for clinically valid training data generation for a decision support system that provides therapy recommendations for rare diseases such as achalasia, while including multi-modal data is presented.

We visited the digital teaching day of the University of Augsburg. Attendees could participate in one of a total of five workshops. Among other things, it was possible to try out such a didactic AR with Medienrealität and eye tracking glasses, while brainstorming how these technologies could be used in teaching. Next, groups of students and teachers were asked to develop ideas together, on how digital teaching could be implemented even further at Augsburg University. These were submitted in digital form, checked for usefulness as well as feasibility. After the evaluation, a prize will be awarded for the best project. Finally, at the Market of Possibilities, various institutions of the university presented their perspective on digital teaching, and what they had to offer.