Reconfigurable Architectures for Electrophysiological Signal Processing

用於電生理信號處理的可重構架構

Student thesis: Doctoral Thesis

View graph of relations

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date9 Aug 2021

Abstract

Electrophysiological signals (sometimes referred to as Biosignals) are electrical activities measured from a living organism. In practice, these signals are frequently used to assess patients' health and to discover bio-physiological mysteries. As a diagnostic procedure, electrophysiological signal analysis is mostly noninvasive. It is widely accessible across the world due to its reasonably low cost. However, as most biosignal processing units are multichannel systems with extensive datasets, conventional computation techniques often fail to offer immediate execution of data processing.

Reconfigurable architecture offers flexible software features with fast parallel computation using the Field Programmable Gate Array (FPGA). This computation technique ensures "Hardware Acceleration," which essentially means the exclusive utilization of hardware resources to expedite computational tasks. This technique is the practice of designing application-specific circuits rather than using general-purpose processors to do signal processing. Because of its low cost and fast parallel computation property, reconfigurable architecture is characteristically suitable for electrophysiological signal processing. In this research, in-depth systematic analyses of four novel FPGA-based reconfigurable architectures are presented. In this works, time-critical and resource-intensive performance issues are addressed, and overall system efficiency improvement is also presented.

An electrocardiogram or ECG records the heart's electrical activity, and it is the most popular noninvasive diagnostic test to determine cardiac health. The first work presents an original ECG compression algorithm with a high-frequency noise reduction protocol. The implemented system can work in real-time and achieve a maximum 90% compression rate without any significant signal distortion. This compression rate is 5% higher than the state-of-the-art hardware implementation. Additionally, this algorithm has an inherent capability of high-frequency noise reduction, making it unique in this field.

An FPGA-based neuron activity extraction unit from the brain signal is described as the second work. This work presents a low-power multichannel active neuron signal extractor suitable for wireless neural interface systems. A novel neural signal extraction algorithm is proposed to achieve the low-power requirement, providing up to 6000X transmission rate reduction considering the input signal. Consequently, this technique offers at least 2X power reduction compared to the state-of-the-art systems.

The third work is about a first-of-its-kind system design for the differential diagnosis of two neuromuscular diseases- neuropathy and myopathy. Electromyography (EMG) is a record of the mussel's electrical activity over a specified period, and it offers a inexpensive method for point-of-care assessment. Extracting only two simple features from the EMG signal and using a simple perceptron machine learning algorithm for the classification, clinical level accuracy is achieved in the implemented prototype.

The last work features a hardware optimized machine learning based Depth of Anesthesia (DOA) measurement framework for mice and its FPGA implementation. The DOA is a measure of consciousness to distinguish whether a patient is suitably anesthetized or not during a surgical procedure. This measurement is critical for both medical and experimental settings where a subject has to undergo the general anesthetic procedure. With the extraction of only two features, the proposed system can classify the state of consciousness with 94% accuracy for a 1 second EEG epoch, leading to a 100% accurate channel prediction after a 7 second run-time on average. Furthermore, this research presents the first-of-its-kind hardware implemented automatic DOA computation system for mice.

All of these researches are compared with state-of-the-art technologies to determine the viability and efficacy of the newly developed systems. A few other potential future research scopes are explored at the end of the thesis.