Department of Bioengineering
http://hdl.handle.net/10106/24280
2024-03-28T18:11:05ZQuantification and identification of neuro-electrophysiological markers and brain network for biomedical applications
http://hdl.handle.net/10106/31746
Quantification and identification of neuro-electrophysiological markers and brain network for biomedical applications
Electroencephalogram (EEG) can detect and monitor neuro-electrophysiological signals in the human brain, including assessing brain function in newborns at risk of neurological injury and healthy adults undergoing intervention with prefrontal transcranial photobiomodulation (tPBM). Moreover, EEG-based brain functional connectivity can be assessed in either resting-state or task-based measurements using graph-theoretical network modeling. However, the management of newborns with mild hypoxic ischemic encephalopathy (HIE) is controversial, and no study has investigated the EEG-based brain network and information flow resulting from HIE. Also, the underlying electrophysiological mechanism of tPBM is still unclear, and further research is needed to determine optimal parameters for tPBM applications. My dissertation targets these gaps to (1) evaluate the potential of predicting neurodevelopmental outcomes of newborns with HIE using the brain state of newborn (BSN) measured within the first day of life; (2) investigate the brain network in newborns with HIE; and (3) compare electrophysiological modulations of the human brain in response to left and right prefrontal tPBM using 800-nm laser.
In Chapter 2, I aimed to predict neurodevelopmental outcomes at two years of age using BSN that was derived from EEG data collected on the first day of life. The results showed that BSN can distinguish normal and HIE cases and has strong correlation with a clinical assessment score (i.e., the concomitant Total Sanart Score). BSN were also differentiated between neonates with normal and abnormal neurodevelopmental outcomes at the age of two years. Additionally, higher BSN values indicate a reduction in the odds of HIE occurrence and of abnormal neurodevelopmental outcomes in global, cognitive, language, and motor skills. The findings confirm that BSN is a sensitive real-time biomarker for monitoring the dynamic progression of neonatal encephalopathy.
In Chapter 3, I targeted the assessment of brain network in newborns with HIE. Based on the first 30 minutes of available clean eight-channel EEG data, I quantified the global brain connectivity parameters in newborns with HIE, followed by comparisons with those from healthy newborns and adults. Furthermore, nodal graphical brain connectivity and region-wise networks were also investigated. The major findings indicate that the neural networks of neonates affected by HIE exhibited a notable reduction of overall efficiency compared to both healthy neonates and adults. However, significant distinctions in these fundamental metrics were not observed between the mild and moderate HIE cohorts, implying the necessity for prompt and efficacious medical intervention even for newborns with mild HIE to mitigate potential adverse outcomes.
In Chapter 4, I explored electrophysiological modulations of the brain in response to left/right prefrontal 800-nm tPBM. Recent literature supports tPBM's capacity to enhance cerebral blood flow and oxygenation and thus to improve cognitive performance. A total of 26 subjects underwent 7-min resting-state 19-channel EEG recordings before and after tPBM/sham stimulation on the left/right forehead, in a single-blind crossover design with randomized sham and tPBM sequences. Global and regional GTA-derived brain networks were assessed and compared between the tPBM and sham conditions. My results indicated site-specific effects of tPBM, with distinct EEG network changes induced by left and right prefrontal tPBM.
2023-08-23T00:00:00ZDESIGN AND DEVELOPMENT OF REAL-TIME SINGLE AND DUAL-WAVELENGTH DCS SYSTEMS TO STUDY SKELETAL MUSCLE BLOOD PERFUSION AND OXYGEN METABOLISM
http://hdl.handle.net/10106/31705
DESIGN AND DEVELOPMENT OF REAL-TIME SINGLE AND DUAL-WAVELENGTH DCS SYSTEMS TO STUDY SKELETAL MUSCLE BLOOD PERFUSION AND OXYGEN METABOLISM
The aim of this study was to utilize Diffused Correlation Spectroscopy (DCS) to design and develop real-time single and dual wavelength systems and assess the physiological impacts of exercises on blood perfusion. The study consisted of 4 parts. Part 1 was to Re-design and develop a real-time single wavelength DCS system, part 2 was to validate this system by a cross validation method with the help of a concurrent study alongside Doppler ultrasound. Part 3 was to design & develop a real-time dual wavelength DCS system and part 4 was to validate the dual wavelength DCS system.
DCS is a developing non-obtrusive procedure to test profound tissue hemodynamics. DCS utilizes time-averaged intensity autocorrelation function for the fluctuations caused by moving scatterers (RBCs) in natural tissue. I introduced a software based autocorrelator framework to finish the securing and handling parts. I led approval considers on an intralipid phantom and human forearm. Both the studies demonstrated smooth decays which aid in showing signs of an improved fitting and subsequently more exact blood flow index (BFI). I appear that the software based autocorrelation framework can be a contrasting option to the traditional equipment based correlators in DCS frameworks with advantages, for example, adaptability in raw photon count information handling, data processing, minimal effort and low cost.
Coordinate nonstop noninvasive estimation of neighborhood muscle blood stream in people stays constrained. Routine estimations of appendage blood stream, for example, Doppler ultrasound or venous impediment plethysmography, measure changes in mass channel stream and don't give provincial data. Near infrared diffused correlation spectroscopy (DCS) is a rising system for non-intrusive estimation of neighborhood muscle blood stream at the microvascular level. To better comprehend the qualities and restrictions of this novel approach, I played out an approval think about by looking at muscle blood stream changes measured by DCS and Doppler ultrasound during a small exercise.
The design and development of a real-time dual Wavelength DCS system consisted of 2 laser sources. An optical switch was used to toggle the light between laser 1 and laser 2. With this dual wavelength system conc. of HbO (Oxy hemoglobin) & conc. of Hb (de-Oxy hemoglobin) can be calculated along with the other parameters like rBFI (relative blood flow index), dOD (delta Optical density) and total intensity of the detected photons. I led validation studies on intralipid phantom and human forearm (arm cuff measurement). Both the studies demonstrated smooth decays which aid in showing signs of an improved fitting for rBFI and conc. of HbO (Oxy hemoglobin) & conc. of Hb (de-Oxy hemoglobin) was also obtained in real-time.
LIGHT REFLECTANCE SPECTROSCOPY FOR THE DETECTION OF POSITIVE SURGICAL MARGIN OF PROSTATE CANCER
http://hdl.handle.net/10106/31701
LIGHT REFLECTANCE SPECTROSCOPY FOR THE DETECTION OF POSITIVE SURGICAL MARGIN OF PROSTATE CANCER
During a radical prostatectomy in the operating room, there exists a certain degree of risks of a surgical positive margin of prostate cancer. Positive surgical margins are defined as cancer that has spread beyond the margin of the prostate and into the surrounding tissues. Surgeons will be able to utilized the concept of Light Reflectance Spectroscopy (LRS) in real-time during surgery to help detect these positive surgical margins and to provide the surgeon a quantitative decision to remove the excess surrounding tissues. This modality would essentially lower the risks of positive surgical margins and prevent the recurrence of cancer, effectively improving the survival of the patient. LRS is highly dependent on the Gleason Score (GS) of the cancer. Based off the 5 and 9 Feature Algorithm, select features of the spectral curve are used to differentiate between normal tissue and cancerous tissue. It is most effective in distinguishing between normal tissue versus high grade cancers such as GS 4+3 and GS 4+4, primarily due to the Gleason Grade 4 pattern. But LRS is less effective in distinguishing lower grade cancers such as GS 6 (3+3) and GS 3+4, primarily due to the Gleason Grade 3 pattern. The 5 Feature Algorithm is the current optimal form of cancer classification algorithm for higher grade cancers, with a Sensitivity and Specificity value of 72.2% and 81.5%, respectively, for the GS 4+4 groups. In addition, there is a question if applying an excess level of pressure onto the surface of the prostate using the optical probe would affect the spectral readings of the LRS measurement. It was found that applying pressures up to ~0.066 N/mm2 does not significantly affect the spectral readings at a peak wavelength value of 630 nm. Also, two new optical probes, a new 370 µm probe and a new 125 µm, were introduced into the study and the penetration depths were calculated. It was found that the new 370 µm probe had a penetration depth range from 1.77 mm to 1.95 mm. The new 125 µm probe had a penetration depth range from 1.06 mm to 1.41 mm.
Development of Mussel Inspired Nanocomposite Adhesives for Biomedical Applications
http://hdl.handle.net/10106/31666
Development of Mussel Inspired Nanocomposite Adhesives for Biomedical Applications
Popular bioadhesives, such as fibrin, cyanoacrylate and albumin–glutaraldehyde-based materials, have been applied for clinical applications in wound healing, drug delivery, and bone and soft tissue engineering; however, their performances are limited by weak adhesion strength and rapid degradation. The long-term goal of this research is to develop a strong tissue adhering nanocomposite for tissue interfacing and wound healing applications. We begin by developing a mussel-inspired, nanocomposite–based, biodegradable tissue adhesive by blending poly (lactic-co-glycolic acid) (PLGA) or N-hydroxysuccinimide modified PLGA nanoparticles (PLGA-NHS) and polydopamine nanoparticles with mussel-inspired polymers. Adhesive strength measurement of the nanocomposites on porcine skin-muscle constructs revealed that the incorporation of nanoparticles significantly enhanced the tissue adhesive strength compared to the mussel-inspired adhesive alone. To further optimize this nanocomposite system, we studied the effects of nanoparticle sizes, concentrations and types as well as types of hydrogel materials including alginate and hyaluronic acid-based materials on the tissue adhesive strengths of the nanocomposites. The nanocomposites made from alginate-based polymers were degradable and cytocompatible in vitro and elicited in vivo minimal inflammatory responses in a rat model, suggesting clinical potential of these nanocomposites as bioadhesives. The Hyaluronic acid-based polymers were found to have the best tissue adhesion at 40% w/v polymer concentration. In addition, among the tested nanoparticles (PLGA, PLGA-NHS, Silica and Polydopamine), the developed polydopamine nanoparticles at 200 nm size and 12.5 % w/v concentrations were found be the most effective in enhancing the adhesion of mussel inspired hydrogels (Alginate-dopamine and Hyaluronic acid-dopamine) with adhesive strengths increasing with further increase in nanoparticle concentrations up to 40 % w/v in the nanocomposite blends of Hyaluronic acid-dopamine polymers. Finally, we developed this optimal nanocomposite adhesive into an antimicrobial tissue adhering degradable system and demonstrate its antimicrobial effectiveness on E. Coli and S. Aureus species of bacteria with potential applications of this nanocomposite for healing of chronic wounds.