Machine learning methods have become a key technology in a variety of fields and have also become an integral part of medical technology. These techniques offer enormous potential in the processing of biomedical data and represent a key competitive advantage in the development of innovative medical devices.
Utilizing modern AI procedures often still depends on strong computing power and corresponding hardware support. This poses particular challenges for medical devices, where inference must be solved in real-time and with high accuracy.
Fraunhofer IMTE offers innovative solutions to apply machine learning algorithms to embedded systems. This paves the way for a new class of biosensors that can process complex multimodal signals and enables new therapeutic and diagnostic approaches. For this purpose, resource-efficient AI methods relying on exploitation of structure and model compression are investigated. At the same time, the use of innovative hardware solutions (e.g. TPUs, FPGAs) is in the focus of current research activities at IMTE.