Graduate Studies, UNL

 

Dissertations and Doctoral Documents from University of Nebraska-Lincoln, 2023–

First Advisor

Steven Barlow

Second Advisor

Gregory Bashford

Degree Name

Doctor of Philosophy (Ph.D.)

Committee Members

Michael Hoffman, Yingying Wang, Yuzhen Zhou

Department

Biomedical Engineering

Date of this Version

2025

Document Type

Dissertation

Citation

A dissertation presented to the faculty of the Graduate College of the University of Nebraska in partial fulfillment of requirements for the degree Doctor of Philosophy (Ph.D.)

Major: Biomedical Engineering

Under the supervision of Professor

Lincoln, Nebraska, December 2025

Comments

Copyright 2025, the author. Used by permission

Abstract

This dissertation establishes a quantitative framework for studying somatosensory processing using functional Near-Infrared Spectroscopy (fNIRS), with a focus on applications in neurorehabilitation. The central aim is to investigate how the brain's hemodynamic response is modulated by the velocity of patterned tactile stimulation. While fNIRS is a promising neuroimaging tool, its data quality is often compromised by physiological noise and motion artifacts, challenging the reliability of its findings and hindering the development of effective, quantifiable neurotherapeutics.

Two preliminary investigations informed the final experimental design. The first pilot study confirmed that pneumotactile stimulation of the hand, during both passive (somatosensory) and active (sensorimotor) tasks, could evoke a significant hemodynamic response. However, it did not reveal a consistent pattern of velocity encoding and was confounded by the high variability associated with the sensorimotor task. A second preliminary study used machine learning classifiers to differentiate the stimulus velocities as reflected in the hemodynamic response. While achieving above-chance accuracy, the performance was inconsistent across participants, underscoring that poor signal quality and experimental complexity were obscuring the underlying neural signals. These initial findings highlighted the critical need for a simplified yet comprehensive paradigm with improved noise mitigation strategies.

To address these challenges, an improved experimental paradigm and a robust data analysis pipeline were developed. Patterned stimuli were administered at four distinct velocities (20, 25, 31, and 39~cm/s) to the hands of 22 right-handed and 3 left-handed neurotypical adults. Crucially, the methodology integrated a novel multimodal recording system, acquiring fNIRS data with short-separation channels alongside a comprehensive suite of auxiliary physiological measurements, including head movements, facial and forearm EMG, respiration, and cardiac pulse. These diverse signals were incorporated as regressors within a General Linear Model (GLM) to systematically isolate the stimulus-evoked cortical activity from systemic interference.

The results validate this comprehensive noise regression strategy, which revealed highly specific and localized brain activation patterns that were otherwise obscured. The primary findings include a robust excitatory response in the contralateral primary somatosensory cortex (S1) and a concurrent inhibitory response in the ipsilateral S1. Furthermore, the analysis demonstrated a velocity-dependent modulation of the hemodynamic response. The comprehensive dataset also allowed for a systematic quantification of how experimental factors and participant characteristics, such as hair type and color, may impact fNIRS signal quality.

In conclusion, this work successfully validates a multi-modal fNIRS-based methodology for the reliable quantification of the cortical response to tactile stimulation. By identifying optimal stimulation parameters and providing a validated framework for mitigating noise, this research offers a neurophysiological basis for designing more effective therapeutic interventions. Moreover, the comprehensive nature of the recordings provides a valuable resource for future investigations.

Advisors: Steven Barlow and Gregory Bashford

Download

Share

COinS