​​Artificial Intelligence has become a central technology for innovation across science, industry, and society. Since its founding in 2018, the Inter-Institutional Doctoral Centre of Applied Computer Science has been committed to advancing AI research through its Special Interest Group Applied Machine Intelligence. The group’s research projects combine technical innovation with a strong focus on real-world applications. Through close collaboration with industry partners, NGOs, and other stakeholders, the researchers aim to generate meaningful technological, economic and societal impact. 

Find out more here.

​​Artificial Intelligence is transforming science – but what does that look like in practice? In this video podcast series, three researchers from the Forschungscampus Mittelhessen offer insights into their work at the intersection of AI and science – from emergency medicine to neuroscience and explainable AI.

​In the first episode, Dr. Kirsten Zantvoort (University of Marburg) investigates how machine learning can safely support clinical decision-making, with a focus on early detection of conditions such as sepsis. In the next episodes Prof. Katharina Dobs (JLU Giessen) explains her use of artificial neural networks to understand how the human brain processes visual information and Prof. Jennifer Hannig (THM) dives into explainable and trustworthy AI.

​Across the three episodes, the cooperation between the Forschungscampus Mittelhessen (FCMH) and the project ”AI for Startups“ at the Technology and Innovation Centre Giessen (TIG) offers a deeper look into the research behind the algorithms and the people who develop them.

​Check out the first episode here.

​​A team of scientists at BAM explores how so-called self-driving labs can be used to synthesize, characterize and optimize nano and advanced materials. Those labs combine the use of machine learning, lab automation, and robotics. A self-driving lab built for speeding up materials discovery is also known as a ”material acceleration platform” (MAP).

​Acceleration is the name of the game: In MAPs, robots can repeat processes and experiments faster than humans can, with high precision. Thus, tests become fully reproducible. Results are automatically examined, evaluated and fed into the next round of trials.

​This is particularly helpful for the development of nano and advanced materials, since they have special requirements regarding the modular hardware building blocks required for their syntheses, purification, and characterization. BAM’s scientists also work on software and hardware tools that help to integrate and control components necessary for producing and characterizing those materials. 

Find out more here and here.

​​At first glance, biology and computer science seem like opposites. But wherever enormous amounts of data are generated from research, progress is hardly possible without digital methods. Bioinformatician Prof. Andreas Keller therefore relies on artificial intelligence (AI). He heads the department “Clinical Bioinformatics” at the Helmholtz Institute for Pharmaceutical Research Saarland (HIPS). In this episode of the podcast “InFact”, produced by the Helmholtz Centre for Infection Research (HZI), he discusses how AI can help us to understand the communication between beneficial and harmful bacteria in our bodies, predict long-term effects of infections and develop new drugs against dangerous pathogens. Find out more about AI-enabled novel drugs here. 

​​At the new Institute for Adaptive Mechanical Systems (IAMS) at the University of Stuttgart, the focus is on a new generation of walking robots and “soft robotics”. The researchers are developing adaptable robots inspired by natural movement patterns and made from innovative soft materials. The applications range from healthcare and industrial production to energy supply. At IAMS, students find an exciting environment that combines basic research with practical development and emphasizes collaboration. ​ 

Find out more here.

​​Conventional image sensors may not be the best option for technological development when it comes to automatically extracting information from image data. The “Learning to Sense” project brings together electrical engineering and computer science. Seven research chairs from the universities of Siegen and Mannheim are working together to develop novel sensors and AI software for tomorrow's cameras, microscopes and smart watches. 

Find out more here.​ 

Rethink everything you know about cybersecurity: the Deggendorf Institute of Technology introduces CAIDAN—an AI-driven intrusion detection and attribution network built for industrial environments. By combining network flow and material flow data, CAIDAN delivers context-aware anomaly detection and transforms scattered alerts into clear, actionable insights.

With a powerful correlation engine, standardized forensic framework, and scalable streaming architecture, the system enables efficient deployment tailored במיוחד for SMEs. CAIDAN not only improves detection accuracy but also enhances situational awareness with interpretable alerts and structured evidence.

The future? Explainable AI, intuitive dashboards, and broader industrial adoption. Discover how CAIDAN can strengthen your cybersecurity strategy today.

Find out more here and here or browse though the current and completed research projects of the Deggendorf Institute of Technology.

Metals have enormous recycling potential—provided their exact composition is determined and impurities are removed. This is precisely where the AlloySort project at TH OWL comes in. Led by Professor Dr. Markus Lange-Hegermann, the team is developing an AI-based solution for the copper and aluminum industries. Until now, no methods have been available to non-destructively analyze heterogeneous recycled materials and determine the exact composition of mixed scrap.

Using PGNAA, AlloySort combines high-resolution sensor data with advanced AI algorithms, including convolutional neural networks, to enable precise material analysis despite high levels of noise. A sorting and conveyor belt demonstrator illustrates how the system can be integrated directly into industrial production processes. Initial results are impressive: the measurement time for identifying metal alloys has been reduced from two hours to only one second. The next project goal is not only to identify alloy types but also to determine their mixing ratios.

Find out more here.​ 

​​Language models based on artificial intelligence (AI) can answer any question, but not always correctly. It would be helpful for users to know how reliable an AI system is. A team at Ruhr University Bochum and TU Dortmund University suggests six dimensions that determine the trustworthiness of a system, regardless of whether the system is made up of individuals, institutions, conventional machines, or AI. The theoretical paper is guided by philosophical concepts of Prof. Albert Newen, Ruhr University Bochum. The technical discussions are delivered by Dr. Carina Newen and Professor Emmanuel Müller from TU Dortmund University. They describe the theoretical framework in the international philosophical journal Topoi.

​Find out more here.​ 

​​Medicines, chemicals, microplastics: even ordinary products can leave a dangerous mark on our environment. “Wastewater is like a mirror of our society,” says Prof. Dr. Matthias Kohl, head of the Data Science for Life Sciences research group at Furtwangen University. Together with PolymerActive GmbH, his team investigates how artificial intelligence can transform microplastic particles into precise “pollutant magnets” that selectively remove harmful substances from water. A new research project led by Prof. Dr. Magnus Schmidt takes this idea even further. The aim is to design the particle surface in a way that certain pollutants bind particularly well, while others hardly attach at all. “Ideally, we want to build a pollutant magnet that only likes certain adhesive partners,” says Schmidt. In this way, AI can help to remove pollutants from water more precisely – contributing to the protection of both the ecosystem and human health. 

Find out more here.

​​Images of segmented cells are of enormous importance for biomedical research and diagnostics. And extremely expensive. Although Artificial Intelligence (AI) is now capable of cell segmentation, it needs to be trained with large quantities of data, which experts must label manually. Within the context of the “Sustainable Life-Cycle of Intelligent Socio-Technical Systems” (SAIL) research project, doctoral candidate Eiram Mahera Sheikh conducts research at HSBI to find out how this process can be sped up efficiently. She thus uses AI – more precisely, Deep Active Learning – as early as in the preparation of training data.​ 

Find out more here.

Scientists are increasingly using artificial intelligence to observe our earth system. This makes it possible, for example, to predict the weather more accurately or warn of natural events such as flooding. However, our planet and the processes taking place on it are too complex for most AI models—especially when they are changing as rapidly as they are due to climate change.  

The research project »AI Generalizability in Non-stationary Environmental Regimes: The Case of Hydro-climatic Extremes (GENAI-X)« is going to develop new AI models that could be used for complex and changing environmental systems. These models will be tested in practical environmental research—particularly in the context of extreme hydro-climatic events such as floods, droughts or landslides. 

Find out more here.