Biosignal Recording with Integrated Circuits

This dissertation presents novel integrated circuits designed to improve signal recording and processing in biomedical applications. The primary focus is on improving low-noise, low-power amplifiers, where a comparison of existing topologies reveals the trade-offs between noise, power consumption and silicon area. Two improved amplifier designs are presented: a three-channel current-reuse amplifier and a split-voltage amplifier. Both designs offer a good noise and efficiency performance compared to the state-of-the-art. Furthermore, this thesis investigates non-invasive electrodes for measuring biopotentials in small worms and for future application on planarians. Two improved electrode configurations are presented: a design with strip lines for longitudinal recordings and a 3D-printed concentric matrix electrode array for multidirectional recordings. The functionality of the non-invasive electrodes is validated by recording action potentials of a redworm, demonstrating their potential for future research on planarians. To improve the signal processing of action potentials and overcome challenges related to high-speed ADCs and digital storage, a nine-channel delay-and-add system for a velocity-selective recording system has been integrated on an ASIC. Finally, this dissertation presents an improved quadratic-integrate-and-fire neuron fabricated in CMOS technology, capable of mimicking various biologically inspired spike patterns. This neuron, in combination with a piezo-FET tactile sensor, shows promising results for future applications in biologically plausible spiking neural networks or as a pacemaker for nerve stimulation. In summary, this dissertation contributes to advancements in biopotential recording and signal preprocessing through improved integrated circuit designs. The presented amplifier designs offer superior performance due to low power consumption and low noise. The use of 3D-printing technology to fabricate electrodes for recording biosignals from small worms is demonstrated and the VSR system as well as the spiking neuron model contribute to advances in signal-to-information conversion and neuromorphic signal processing. These findings pave the way for miniaturized, low-power, and application-specific biopotential recording systems.

Level thesis:degree_level

thesis.doctoral

Grantor dc:publisher

Christian-Albrechts-Universität zu Kiel

Year

2025

Author dc:creator

• Simmich, Sebastian

Contributors dc:contributor

• Rieger, Robert
• Kohlstedt, Hermann
• Demosthenous, Andreas

• ASIC
• Biosignal Recording
• Analog
• Efficient
• Amplifier
• Integrated Circuits

Repository record source_url

https://macau.uni-kiel.de/receive/macau_mods_00008353

OAI identifier oai:identifier

oai:macau.uni-kiel.de:macau_mods_00008353