CaltechTHESIS
  A Caltech Library Service

Electrical Impedance Spectroscopy-derived 3D Conductivity Tomography for Atherosclerosis Detection

Citation

Huang, Zi-Yu (2022) Electrical Impedance Spectroscopy-derived 3D Conductivity Tomography for Atherosclerosis Detection. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/n0md-h379. https://resolver.caltech.edu/CaltechTHESIS:12202021-172349636

Abstract

Electrical impedance tomography (EIT) utilizes voltage/current data measured from the surface of interest to reconstruct the electrical conductivity distribution. This results in a noninvasive medical imaging procedure with many applications. Some examples would be: lung ventilation monitoring, breast cancer detection, and fatty liver detection. Non-alcoholic fatty liver disease (NAFLD) is one of the most common causes of cardiometabolic diseases in overweight individuals. The gold standard for NAFLD diagnosis is a liver biopsy which is a risky and invasive procedure. A non-invasive and cost effective method to detect fatty liver is an important unmet clinical need. Due to the distinct electrical properties of fatty tissue versus normal tissue, EIT can be applied to detect the fat infiltrate in the liver. We conducted EIT measurements and reconstructions on 19 subjects where the fat infiltrate was validated by MRI proton-density fat fraction (PDFF). The liver EIT conductivity was shown to be inversely correlated with MRI PDFF, demonstrating the ability of EIT to detect fatty infiltrate in the liver.

This thesis also extends the EIT reconstruction to detect atherosclerosis, which is a build-up of fatty tissue in the arteries (plaque). Some plaques are prone to rupture and the current gold standard has a false negative rate of 20 % when distinguishing between vulnerable plaque and stable plaque. We sought to use EIT to detect the fatty content (mainly oxidize LDL) inside these vulnerable plaques. Therefore, the reconstruction method was modified into an outward setting that can measure from the inner surface of interest. Ex vivo experiments have demonstrated the ability to detect the location of fatty tissue in swine aorta. This technique has the potential to detect vulnerable plaque. However, the dimension of the device and the required electrode number limits the application from in vivo animal artery experiments.

Finally EIS-derived EIT, a new method we proposed, utilizes impedance values at a fixed frequency to solve for the conductivity distribution. This approach circumvents the mathematically ill-posed problem found when performing traditional EIT methods. We designed a 6-point EIS electrode array that was circumferentially configured to a balloon catheter and deployed in Yorkshire mini-pigs with induced stenosis in the right carotid artery. The EIS spectra demonstrated an elevated impedance in the right carotid arteries and the EIS-derived EIT mappings were reconstructed. The low conductivity regions in the EIS-derived EIT mappings were correlated with the positive E06 immunostaining for oxLDL-laden regions. Thus, we establish the capability of 3D EIS-derived EIT to detect oxLDL-laden arterial walls with translational implication to predict metabolically active plaques prone to acute coronary syndromes.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Electrical impedance spectroscopy; Electrical impedance tomography; Atherosclerosis
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Medical Engineering
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Tai, Yu-Chong
Thesis Committee:
  • Wang, Lihong (chair)
  • Emami, Azita
  • Gao, Wei
  • Tai, Yu-Chong
Defense Date:9 December 2021
Record Number:CaltechTHESIS:12202021-172349636
Persistent URL:https://resolver.caltech.edu/CaltechTHESIS:12202021-172349636
DOI:10.7907/n0md-h379
Related URLs:
URLURL TypeDescription
https://doi.org/10.1038/s41598-021-99132-zDOIArticle adapted for Chapter 2 and 3.
https://doi.org/10.1109/TBME.2021.3104300DOIArticle adapted for Chapter 4.
https://doi.org/10.1016/j.snb.2021.131152DOIArticle adapted for Chapter 5 and 6.
ORCID:
AuthorORCID
Huang, Zi-Yu0000-0001-5998-3097
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:14454
Collection:CaltechTHESIS
Deposited By: Zi Yu Huang
Deposited On:22 Dec 2021 00:43
Last Modified:03 Jan 2022 17:20

Thesis Files

[img] PDF - Final Version
See Usage Policy.

198MB

Repository Staff Only: item control page