Westworld, I, Robot, Frankenstein – Man-made artificial intelligence has fascinated and inspired writers and filmmakers for decades, if not centuries. Hopes and fears are stirred up, utopias and dystopias are described and the legend of an artificial, more or less free-thinking, feeling creature is created. This fiction creates a magic around artificial intelligence that has little to do with the real possibilities and opportunities. When we talk about artificial intelligence today, we mean, for example, the independent classification of photos by the algorithm of a search engine operator, but also the translation of texts and more intelligent spelling corrections that take the grammar of a language into account and can also provide writing tips. It’s about recognizing patterns in complex data sets and letting algorithms do the hard work. To do this, large, unsorted amounts of data must be automatically processed, classified and, if necessary, transformed.
The chances of AI are that people can be accompanied by artificially intelligent “assistants” in repetitive, uncreative and possibly error-prone work. The aim is not to replace people, but rather to strengthen their abilities. E.g. to improve the language level with only mediocre language skills through AI-based writing assistants,
explained Nane Kratzke, co-speaker of the artificial intelligence in application (KIA) section of the Technical University (TH) Lübeck. At the TH Lübeck, methods of AI and machine learning are used in a wide range of application fields such as acoustics, cloud computing, data science, e-learning, hydrology, IT security, communication systems, medical technology, predictive maintenance, human-computer interaction or control technology applied.
In addition, the university is committed to sharing and making available knowledge related to AI via various platforms and networks.
At the THL, work is currently being done on applications such as
- AI to reduce study dropouts,
- for patient diagnostics and
- can be used to optimize communication networks.
Learning with AI
“We could now predict dropouts at universities with a relatively good hit rate,” he said Monique Janneck, head of Institute for Interactive Systems of the TH Lübeck, “however, the information required for this is often not accessible to course directors, lecturers or students.” In the project DiSEA want the researchers of the TH Lübeck and the Berlin University of Technology therefore draw conclusions about the study success from the user behavior of the students in a learning management system. Machine learning methods are used for this. The models developed from this are to be used in early warning systems in order to give students feedback on their learning behavior and to be able to advise students at risk more specifically. The students’ consent to the evaluation of the usage data is essential: “As in all AI projects, the quality of the application stands and falls with the quality of the data on which it is based,” explains project manager Janneck.
We therefore rely on students to give their consent and involve them closely in the development process to promote acceptance and develop solutions that offer real added value for the students.
Patient diagnostics with AI
In rural areas, it is often difficult to reach the nearest specialist. Patients with restricted mobility in particular have to make a lot of effort to reach the nearest practice. At the same time, a quick diagnosis close to the patient is often necessary in medicine. Therefore, researchers from the TH Lübeck and the University of Luebeck how general practitioners can use smartphones to diagnose eye diseases.
Current smartphones are ideal for this task due to their widespread use and the already built-in sensors and computing capacity, but are currently only used sporadically and rudimentarily,
explained Horst Hellbrueck, head of the competence center CoSA the TH Lübeck the idea of the project PASBADIA. The high-quality cameras together with the light sources built into smartphones (LED flash) allow the implementation of optical diagnostic methods from the field of classic devices. With the appropriate attachments and software applications, the family doctor can use the smartphone to carry out basic eye examinations and thus save the patient the trip to the ophthalmologist. The artificial intelligence helps to interpret the measurement data correctly and to show treatment options.
AI for 5G wireless networks
Well-functioning radio networks are a key factor in the infrastructure of a modern country. the Innovation report of the Smart Service World shows that mobile and dynamic radio networks are being used steadily in industry and medical care. In order for this to work optimally, the available communication resources such as the frequency ranges and the bandwidth of 5G networks must be dynamically allocated as needed. With the help of AI, the researchers, led by Prof. Hellbrück, can predict who will need which resources. In this way, they support an optimal distribution of resources and make the network more reliable. The effectiveness of this support is being demonstrated in a test field in Schleswig-Holstein.
Sharing knowledge
As a public university, it is particularly important to TH Lübeck to let citizens, companies and institutions share their knowledge about AI. That is why there are various projects at the university to support this transfer of knowledge. in the AI Transfer Hub Schleswig-Holstein should together with the WTSH in particular, the AI knowledge transfer for small and medium-sized enterprises (SMEs). The application of AI technologies is considered an important factor for future innovation and competitiveness. Therefore, the regional companies should benefit from the knowledge of the university. The establishment of an innovative network in the field of AI transfer bundles the AI competence of the state and makes it visible as a beacon beyond the national borders. In addition, a low-threshold digital qualification offer for SMEs and their employees to get started with the topic of artificial intelligence and its possible applications was created.
In addition, the university is developing the Future Skills education platform for Schleswig-Holstein. The digital curriculum focuses in particular on basic training in artificial intelligence (AI), which includes both university content and external content.
AI Week in Lübeck
From the November 15-19, 2021 the “Week of AI” will take place in Lübeck, on November 25, 2021 the next edition of “BioMedTec Ideas”. Here is the programme.