Doctoral Theses

Permanent link for this collection

https://hdl.handle.net/10292/4
The Doctoral Theses collection contains digital copies of AUT doctoral theses deposited with the Library since 2004 and made available open access. All theses for doctorates awarded from 2007 onwards are required to be deposited in Tuwhera Open Theses unless subject to an embargo.

For theses submitted prior to 2007, open access was not mandatory, so only those theses for which the author has given consent are available in Tuwhera Open Theses. Where consent for open access has not been provided, the thesis is usually recorded in the AUT Library catalogue where the full text, if available, may be accessed with an AUT password. Other people should request an Interlibrary Loan through their library.

Browse

Browsing Doctoral Theses by Supervisor "Anand, Gautam"

Now showing 1 - 2 of 2
  • Item
    Compensating for Signal Saturation Due to Electrostatic Noise in Electroencephalogram (EEG) Acquisition
    (Auckland University of Technology, 2024) Heidarian, Mahdi
    The electroencephalogram (EEG) is a non-invasive tool for monitoring the electrical activities of the brain for different applications from clinical and neurological research to brain-computer interface (BCI) platforms. EEG can be acquired by different types of electrodes like wet, dry contact, and capacitive electrodes. Although wet/gel electrodes are the gold standard for EEG recording in clinical applications and for short periods of tests, they are not a preferred option during continuous daily activities and for wearable gadgets. Shortcomings of gel electrodes and the critical role of prompt response after brain stroke, have inspired researchers to design dry electrodes and try to make them more compatible with wearable EEG measurement systems. Dry EEG electrodes can record the signals through galvanic contact to the scalp skin or capacitively without a direct connection to the skin. The electrode-tissue impedance (ETI) in both types of dry EEG sensory systems is highly associated with the hair-air domain which can cause erroneous measurements due to unknown impedance characteristics of hair, the potential of having heavy motion artefacts in wearable measurement systems, and hair’s static electricity discharge through the system. These effects can potentially lead to long-term blockage and/or heavy baseline drifts in high-gain front-end dry EEG electrode circuits. The main aim of this study is to evaluate and mitigate the effect of electrostatic noise on dry EEG electrodes. To achieve this, a simulation of an active band-pass filter employing an operational amplifier (op-amp) was conducted. This filter is designed to operate within the EEG frequency band, between 0.72 Hz and 72 Hz. Recognizing the weak nature of EEG signals on the scalp, a gain of 220 was selected. In addition, a simplified lumped model representing the hair-air domain was integrated into the input stage. This model was applied to both contact and non-contact dry electrodes. To mimic electrostatic noise, the human-body model (HBM) based on the MIL-STD 833 standard was employed. This choice aimed to replicate real-world conditions, ensuring a comprehensive simulation. After evaluating the designed amplifiers in normal working conditions, simulations were completed to observe the progression from mild disruption to complete signal loss as the injected charge increased. Furthermore, the relationship between the physical properties of the hair-air domain and the duration of the signal disruption was evaluated. Moreover, to validate the simulation outcomes, the printed circuit board (PCB) layout of the simulated circuits was manufactured and tested. The experiment results are aligned with the simulation results. The electrostatic noise was injected into the circuit using a manufactured electrostatic discharge (ESD) simulator that is capable of injecting charge with MIL-STD 833 standard characteristics. Once the simulation results of destructive effects of the electrostatic noise were confirmed, compensation measures were devised and implemented on the PCBs to mitigate the impact of electrostatic charges. In dry contact electrodes, a discharge pathway was established to facilitate the dissipation of the charge. For dry non-contact electrodes, the compensation approach encompassed a detection/control unit capable of detecting saturation in the amplification stage, along with a Howland current pump, which, in combination, served to elevate the input baseline and expedite the return to normal operational conditions within a shorter timeframe. The effectiveness of these compensation strategies was subjected to testing and validation using the developed PCB layouts, and the ESD simulator. Conclusively, a homogenous human head phantom made of gelatine and covered with genuine Remy human hair was used. This setup served to assess the electrodes’ capability to replicate human hair conditions. To introduce electrostatic charges, an IMCS 2600 ESD gun was employed. Subsequently, the dry contact EEG electrode, both with and without compensation techniques, was employed to capture signals and appraise the effectiveness of the designed dissipation path. During the experiments, the signal loss occurred on the contacted EEG electrodes without the compensation strategy, although the duration was comparatively shorter than observed in the PCB circuit experiments. The compensation strategy proved effective, eliminating signal loss entirely by providing a discharge path, as confirmed by the simulation results. The results of this PhD research introduce a novel method to decrease the signal blockage duration in active dry EEG electrodes, rendering them more viable for integration into wearable biopotential measurement devices.
  • Item
    Estimation of Arterial Diameter Using Bioimpedance Spectroscopy on a Human Wrist Phantom
    (Auckland University of Technology, 2022) Yu, Yang
    Background: Blood pressure measurement (BPM) is a well-known clinical method to monitor cardiovascular function, and it is also a reliable predictor of death and cardiovascular disease. Since the beginning of this century, a type of cuff-less continuous BPM has been investigated based on the strong relationship between pulse wave propagation (i.e., pulse wave velocity, pulse transit time, and pulse arrival time) and blood pressure (BP). A comprehensive review was undertaken to explore the limitations of the existing pulse wave propagation method (PWPM), including the techniques, mathematical models, and clinical protocols. It was found that the lack of absolute arterial diameter information in previous studies might limit the performance of existing PWPM because the arterial diameter change has been proved to be one significant factor in BP change from both theoretical and experimental perspectives. Hemodynamic monitoring is concerned with the dynamic blood flow within the human circulatory system. Bio-impedance measurement (BIM) can sense the physical and electrochemical processes in human tissues and hence can monitor various physiological variations. In the context of ambulatory hemodynamic monitoring, there has been interest in portable BIM for both diagnostic and research purposes by placing the electrodes on human extremities, such as the wrist. The radial artery is a common location to detect pulsatile blood due to the thin surrounding tissue layers, which has been used for pulse wave propagation determination. When a pulse wave arrives, the amount of blood inside the artery increases and the measured impedance decreases because of the higher conductivity of the blood. Therefore, BIM at the wrist is a promising technique to estimate the arterial diameter, offering an improvement in PWPM for cuffless BPM in the future. Objectives: The main objective of this research is to improve the accuracy of arterial diameter estimation from bio-impedance signals by reaching a consensus between observed impedance values (from computational simulation and phantom experiments) and mathematical modelling. This thesis focuses on the effects of different electrode configurations on current density and electric field (E-field) distribution within the wrist, aiming to achieve a reasonably uniform E-field distribution such that the cross-sectional area changes of the blood can be estimated more accurately. Method: Finite element analysis was performed on a 3D human wrist segment containing fat, muscle, and a blood-filled radial artery. Then, the skin layer, bones and a contralateral blood-filled ulnar artery were stepwise added, helping to understand the dielectric response of multi-tissues and blood flow in the 𝛽 -dispersion band (from 1 kHz to 100 MHz), the current distribution throughout the wrist, and the optimisation of electrode configurations for arterial pulse sensing. Two wrist phantoms were fabricated to verify the simulation results from both one-artery model and two-arteries models. Each wrist phantom contained two components: (1) the surrounding tissue simulant was fabricated by mixing 20 wt.% gelatine power with 0.017 M sodium chloride (NaCl) solution, (2) the conductive blood was simulated using 0.08 M NaCl solution. The blood-filled artery was constricted by a commercial desktop injection pump, and the impedance change was synchronously measured using the multi- frequency impedance analyser. Main results: The simulation results indicated the promising abilities of band electrode method to generate a more uniform current distribution than the traditional spot electrode approach. Both simulation and phantom experimental results demonstrated that a longer spacing between current-carrying (CC) electrodes with shorter spacing between pick-up (PU) electrodes in the middle region can sense a more uniform E-field, engendering a more accurate arterial diameter estimation. For the one-artery model, the arterial diameter could be accurately estimated with an average percent error of less than 1% in both simulation and phantom experiments. For two synchronously pulsatile arteries, the band electrode configuration exhibited a significantly higher accuracy in sensing overall blood volume change throughout the measured region. Conclusion: In summary, this thesis contributes to the accurate quantification of arterial diameter-dependent impedance variance by investigating the effect of electrode configuration. A promising band electrode configuration was developed for more accurate arterial diameter estimation from the numerical simulation and tissue phantom perspectives. More accurate arterial diameter estimation via BIM could further improve the performance of existing PWPM and cuffless BPM in the future.
Theses are protected by the Copyright Act 1994 (New Zealand). The thesis may be consulted by you, provided you comply with the provisions of the Act and the following conditions of use:
  • Any use you make of these documents or images must be for research or private study purposes only, and you may not make them available to any other person.
  • Authors control the copyright of their thesis. You will recognise the author’s right to be identified as the author of the thesis, and due acknowledgement will be made to the author where appropriate.
  • You will obtain the author’s permission before publishing any material from the thesis.