AI-based wound monitoring

The KISMADI (Kognitive Intelligenz für die Medizinische Diagnostik) project is developing a wound dressing equipped with wireless sensors that measure and transmit various parameters of a wound, including temperature, pH value, conductivity, and fluid production (exudate quantity). An artificial intelligence system, programmed and trained by Fraunhofer SYMILA, uses the data to evaluate and classify the wound. The aim is to predict the further healing process and to provide treatment recommendations so that the wound is highly likely to heal.
Fraunhofer SYMILA has developed a Double Digital Twin for this purpose. The first twin maps doctors' knowledge about wounds and allows data models to be derived. This allows us to generate synthetic data for training deep neural networks. The models created in this way are mapped in the second digital twin, which generates initial prognoses of the wound healing process and is gradually improved by measurement data from the wound during the healing process. This approach makes it possible to arrive at a valid model more quickly. It can easily be transferred to other fields.
The Fraunhofer Institute for Silicate Research ISC is developing the silica-based wound dressing. The Bioanalytics and Bioprocesses department of the Fraunhofer Institute for Cell Therapy and Immunology IZI-BB is enhancing the dressing for a longer service life. The Fraunhofer Institute for Reliability and Microintegration IZM is developing and integrating the sensors.
Fraunhofer FIT is responsible for creating the AI models and integrating the software, for data storage and overall data management.
The KISMADI project is funded by the German Federal Ministry of Education and Research; it is coordinated by the Fraunhofer Center for Digital Diagnostics.
Your benefits
|