The project at the Naturwissenschaftliches und Medizinisches Institut (NMI) aimed to create flexible neural interfaces capable of high‑resolution recording and selective stimulation of the visceral nervous tissue. The interfaces were designed to be biocompatible, mechanically compliant, and electrically stable, with sufficient isolation and low noise to allow continuous operation in vivo. The target tissue was the visceral plexus located caudal to the aortic bifurcation in the rat, a region chosen because the nerve fibers are arranged in a slightly fan‑shaped configuration that facilitates selective contact.
A flexible micro‑electrode array (MEA) was fabricated on an 8 µm thick polyimide substrate. The array contains 32 recording sites, two reference electrodes, and two ground electrodes, all made of titanium nitride (TiN). Three different electrode diameters (30 µm, 60 µm, and 100 µm) were produced to evaluate the trade‑off between signal‑to‑noise ratio and charge injection capacity. Biphasic current pulses with phase durations of 200–500 µs, amplitudes up to 800 µA, and an inter‑stimulus interval of 4 s were applied through a bipolar stimulating electrode supplied by the partner Inomed. In the initial cadaver experiments, the MEA was mechanically reinforced with a polyimide ring and placed directly on the exposed plexus. Recordings from all 32 sites showed a similar temporal profile. When stimulation currents exceeded 500 µA, a post‑stimulus potential was observed on every electrode, which was attributed to muscle activity rather than neural firing. The absence of detectable neural activity was attributed to the long time elapsed between animal death and recording, which likely caused degradation of the nerve tissue.
Based on these results, the electrode layout was refined to match the rat anatomy more closely. The optimized design was then tested in fresh tissue preparations, confirming that the array could be positioned with minimal tissue distortion and that the recording quality remained high. The next phase of the project will involve adapting the design for large‑animal studies at the University Hospital Tübingen. In parallel, the NMI team is working on artifact‑free signal acquisition. Collaboration with the partner MCS will provide low‑noise, high‑input‑range amplifiers that suppress stimulation artifacts, enabling simultaneous stimulation and recording.
The project was carried out from 1 November 2018 to 31 October 2021 under the NEPTUN funding program (grant 13GW0271C). The NMI project leader was Peter Jones. The research was conducted in close cooperation with the University of Tübingen (UKT), Inomed, and MCS. UKT supplied the surgical expertise and animal models, Inomed provided the stimulating electrodes and assisted with signal analysis, while MCS contributed the amplifier technology and integration support. Throughout the project, NMI staff trained surgeons and participated in operations to ensure that the interface design met clinical requirements. The collaboration also involved joint development of stimulation protocols and data‑analysis models to identify optimal stimulation patterns from the recorded signals. The combined effort of these partners has produced a flexible neural interface platform that is ready for further validation in large‑animal experiments and, ultimately, for translation to human therapeutic applications.
