Purdue University Graduate School
A.Kirby_Thesis.pdf (2.62 MB)

Time and Frequency Domain Analysis of Physiological Features During Autonomic Dysreflexia After Spinal Cord Injury

Download (2.62 MB)
posted on 2022-07-22, 20:12 authored by Ana K KirbyAna K Kirby


Persons with a spinal cord injury (SCI) often suffer from secondary complications including the dysfunction of the autonomic nervous system below the level of injury. For persons with a SCI at or above T6, autonomic dysreflexia (AD) may be triggered by noxious stimulation below the level of injury causing rapid sympathetic hyperactivation, leading to paroxysmal hypertension. If AD is not recognized and managed promptly, this increase in blood pressure can lead to stroke, organ damage, and/or death. Currently, AD is only detected in clinical settings through continuous blood pressure monitoring. Recent studies have revealed that rapid detection of AD is possible by using extracted features from electrocardiogram (ECG) data collected non-invasively and applying a five-layer neural network.

This project focuses on further characterization of physiological responses before and during AD to detect the overreaction of sympathetic nerve activity prior to the detrimental increase in hypertension. Using a rat model with implanted telemetry and noninvasive sensors, an acclimation protocol was developed to minimize noise and motion artifacts during data collection. We induced AD in a controlled manner using colorectal distention (CRD). We analyzed skin nerve activity (SKNA) and heart rate variability parameters in the time and frequency domain to improve the non-invasive detection of AD. A four-week acclimation protocol exposed a minimal increase in sympathetic activity during experimentation despite long periods of restraint. Results indicated an increase in SKNA features occurred about 18 seconds before the gold standard increase in blood pressure. Additionally, integrated SKNA features in the frequency domain quantified nerve activity and low frequency components were found to be dominant during AD, providing another parameter that could be included in an AD detection system to improve accuracy. In humans, SKNA may be used to alert patients of the onset of AD, allowing caretakers to respond quickly and make necessary changes to decrease the severe effects of AD.


DOD CDMRP Spinal Cord Injury Research Program W81XWH-20-1-0725


Degree Type

  • Master of Science


  • Biomedical Engineering

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Bradley S. Duerstock

Additional Committee Member 2

Thomas H. Everett IV

Additional Committee Member 3

Riyi Shi

Usage metrics



    Ref. manager