Artificial Intelligence offers great opportunities for the treatment of illnesses but also for the development of new business fields in the health economy. North German clinics, universities, research facilities, and companies have therefore joined together for the KI-SIGS project so that they can network and begin their first joint projects. The Center for Computing Technologies (TZI) at the University of Bremen is involved in two sub-projects that are working towards the superordinate aim of creating a permanent network named KI-Space. The sub-projects in Bremen deal with the treatment of eye illnesses and the support of movement therapies.
Intelligent Image Assessment for Ophthalmology
Some of the involved TZI scientists are investigating intelligent image assessment in ophthalmology. The Human-Computer Interaction working group, led by Professor Johannes Schöning, is supporting the development of an AI platform for the care of patients suffering with illnesses such as age-related macular degeneration (AMD). By means of AI-based methods of image analysis and the automatic assessment of 3D image data, a significant improvement in therapy of AMD patients is to be attained. Images of patients taken at home on smartphones can then be used for the automatic recognition of a deterioration of the condition in order to determine the optimum time of treatment and to reduce the number of required doctor’s visits.
The TZI team is developing an app that makes a more precise positioning of a smartphone in front of an eye possible and thus ensures a better photo quality. A decisive factor is the simple, intuitive, and efficient handling by the patients, of which most are of an older age. “With our work, we want to identify problems with user experience when smartphone photos of eyes are being taken,” explains Schöning. “We wish to find out how we can solve said problems and create a user-friendly application that makes it easily possible to take photos that are suitable for diagnostic purposes.”
Movement Training: Recognition of Body Posture Despite Covered Body Parts
The second TZI project focusses on the support of movement therapies. Specially directed movement training is required for many therapy and rehabilitation measures, as well as a preventive measure for the aged. As it is not always possible to provide individual instructions given by therapeutic specialists, especially when exercises are carried out in one’s own home, assistant systems can provide valuable support. For example, robot systems can give the user individual feedback regarding the movements carried out and can thus enable autonomous training.
“However, the reliable 3D recognition of body posture is a challenge that needs to be mastered in order for a number of physiotherapeutic and interactive health applications to be realized,” explains Professor Rainer Malaka, managing director of TZI and head of the Digital Media Lab. “Individual body parts are often covered during more complex chains of movement, for example exercises when one is squatting or lying down. AI-based systems can aid the correct interpretation of the movements.”
TZI Successful with Further Project in BMWi AI Competition
The Federal Ministry for Economic Affairs and Energy (BMWi) is funding the KI-SIGS project, which is being coordinated by UniTransferKlinik GmbH (Lübeck), as part of the Artificial Intelligence as a Driver for Economically-Relevant Ecosystems (“Künstliche Intelligenz als Treiber für volkswirtschaftlich relevante Ökosysteme”) with around ten million euros until 2023. The TZI at the University of Bremen is already a part of the same funding program with the Knowledge4Retail project, in which AI systems are being developed for retail.
Center for Computing Technologies (TZI)
University of Bremen
Phone: +49 (0)171 5305119