ADVANCED INTRACORTICAL NEURAL PROBES WITH ELECTRONIC DEPTH CONTROL
Prof. Dr. Oliver Paul (Contact PI)
The interdisciplinary project Advanced EDC develops the next generation of implantable CMOS-based brain-computer interfaces for high-density intracortical recordings. The technological development of these devices aims at a pronounced reduction of probe dimensions to minimize tissue damage while the available channel count is increased. This is achieved by a hierarchical electrode addressing scheme and circuitry for a local low noise amplification (LNA) and analog-to-digital conversion (ADC) integrated directly on the slender probe shafts. Advanced algorithms based on machine learning approaches are developed and validated to analyze the recorded data, extract relevant information from the neural signals and use this information for planning the optimal probe reconfiguration. These algorithms are integrated in a separate chip bringing data analysis and electrode reconfiguration closer to the implanted probes, as required for in vivo experiments with freely behaving animals.
Development of a CMOS-based neural probe with hierarchical addressing scheme for 1600 recording sites and 32 output channels, smallest ever published closed-loop LNA (foot-print: 50×80 µm2) integrated in probe shafts next to the recording sites, development of a learning algorithm for optimal channel selection using a multi-armed bandit framework to reduce signal redundancy, system-on-chip implementation of the learning algorithm benefiting from highly parallelized FPGA computation.
Kuhl M, Manoli Y (2015) A 0.01 mm² Fully-Differential 2-Stage Amplifier with Reference-Free CMFB Using an Architecture-Switching-Scheme for Bandwidth Variation. Proceedings of the European Solid-State Circuits Conference (ESSCIRC), pp: 287-290.