UCLA

ABSTRACT OF THE THESIS

A Programmable Wireless Single Channel Neural Interface with Artifact Cancellation Capability

by

Yu NongMaster of Science in Electrical and Computer Engineering University of California, Los Angeles, 2022 Professor Wentai Liu, Chair

In the past decades, the field of neural interface has gained significant amount of attention and advancement. However, some desirable powerful features have not been designed in the current neural interface devices because of technical challenges. First, most current neural interface devices use wire communication to transfer collected data and receive external commands. While wire communication provides better signal quality and data rate, wireless communication enables the free movement of experiment target in chronic studies. Secondly, stimulation artifact becomes a major roadblock for the development of bidirectional neural interfaces. This problem can be further divided into two parts, the saturation of amplifier caused by large-amplitude artifact and the distortion of neural signals. Thirdly, for rodent-based chronic studies, the size and weight become another layer of constraints given the size of experiment animal. Fourthly, various neuromodulation applications usually require different stimulation protocols, thus the stimulation waveforms generated by the interface must be versatile and programmable to support a broad set of applications – a one size fits all concept. To address the aforementioned features, this thesis introduces a low-power mobile, lightweight, and wireless single channel stimulation and recording system with real-time artifact cancellation capability. The proposed system consists of a remote/implanted device, a wireless communication adapter, and a graphical user interface (GUI) on a laptop/desktop. User can execute commands (e.g., toggle stimulation/recording, adjust stimulation parameters, enable artifact cancellation) to the implanted/remote device through a GUI in real-time. Bluetooth Low Energy (BLE) is used as the wireless communication protocol between the implanted/remote device and the adapter. An adapter is used to accommodate the speed limitation of BLE on laptop/desktop. The size of the current prototype is approximately 8cm�3cm and has the potential to shrink down to 2cm�1.75cm. Real-time artifact cancellation capability is realized through the combination template subtraction followed by Hampel filter. The functionality and robustness of the system has been validated in both in-vitro and in-vivo experiments.