Scientists at the Chair of Optoelectronics at TU Dresden have succeeded in developing a biocompatible implantable AI platform: it recognizes which patterns in biological signals such as heartbeats are healthy and which changes are to be classified as pathological, even without the involvement of doctors.
The research team led by Prof. Karl Leo, Dr. Hans Kleemann and Matteo Cucchi is pursuing an approach for the real-time classification of biosignals based on a biocompatible AI chip. The scientists use polymer-based fiber networks that are structurally similar to the human brain and enable the neuromorphic AI principle of reservoir computing.
The polymer fibers are randomly arranged and form a so-called "recurrent network". This allows data to be processed in a similar way to the human brain. Because these networks work non-linearly, even the smallest signal changes can be amplified, which doctors often find difficult to assess. However, the non-linear transformation with the help of the polymer network makes this possible without any problems. In tests, the AI was able to distinguish healthy heartbeats from three common arrhythmias with an accuracy of 88%. The polymer network consumed less energy than a pacemaker.
There are many possible uses for implantable AI systems: for example, they could be used to monitor cardiac arrhythmias or complications after operations and report them to doctors and patients via smartphone. This enables rapid medical assistance.
With this approach, it is possible to develop further intelligent systems in the future that can help save human lives.
