Edge AI: smart sensor technology for medicine and the environment

A research consortium in Brandenburg is developing advanced edge AI technologies aimed at improving sensor systems for medical technology, industrial electronics, and environmental monitoring. Under the project name InSeKT (“Intelligent Sensor Edge Technologies”), the Technical University of Applied Sciences Wildau, the Leibniz Institute for High Performance Microelectronics (IHP), and the Fraunhofer Institute for Photonic Microsystems (IPMS) are collaborating to create compact sensor systems that process data directly at the source — the “edge” of IT networks. 

In contrast to today’s widespread reliance on centralized cloud computing, the new approach focuses on decentralized data processing. Large volumes of data no longer need to be transmitted over long distances, which reduces latency, minimizes energy use, and significantly enhances data security. For applications in medical technology, industrial automation, and environmental analysis, these capabilities are increasingly essential. 

Fraunhofer IPMS contributes its long-standing expertise in microsystems engineering to the project. At its Cottbus-based Integrated Silicon Systems branch, researchers are developing miniaturized MEMS sensors (micro-electromechanical systems) that not only collect data but also process it directly within the device. This approach supports: 

• More compact sensor formats 

• Cost reduction through minimized hardware requirements 

• Improved reliability of systems 

• Scalability across multiple application areas 

1. Gas analysis using ion mobility spectrometers (IMS)
   A new demonstrator system is being developed based on FAIMS (field asymmetric ion mobility spectrometry). With flexible electrode spacing and high sensitivity, this sensor can detect ionizable gases even at very low concentrations. Possible applications include environmental monitoring and security systems.
2. Photodetectors for near-infrared light
   These detectors are used in material analysis and recycling, and are capable of analyzing materials through packaging. The aim is to improve the performance and cost-efficiency of Al-TiN-Si Schottky detectors by enhancing their sensitivity using cylindrical-pyramidal structures and affordable materials.
3. CMUTs for medical ultrasound imaging
   Capacitive micromechanical ultrasonic transducers (CMUTs) offer high resolution for imaging applications. These are particularly promising in medical diagnostics, including movement analysis and blood glucose monitoring.

“Later on, it will be possible to carry out very precise analyses of hand movements using an ultrasound signal based on that of bats, as well as measuring blood sugar using ultrasound,” says Dr. Sebastian Meyer of Fraunhofer IPMS. 

COMPAMED-tradefair.com; Source: Fraunhofer Institute for Photonic Microsystems (IPMS)