Browsing by Author "Harris, Ashley"

Now showing 1 - 9 of 9
  • Thumbnail Image
    ItemOpen Access
    A voxel-level approach to brain age prediction: A quantitative method to assess regional brain aging
    (2023-12-05) Gianchandani, Neha; Souza, Roberto; MacDonald, Ethan; Bayat, Sayeh; Pike, Bruce; Harris, Ashley; Tan, Benjamin
    Global brain age has been used as an effective biomarker to study the correlation between brain aging and neurological disorders. However, brain aging is a regional phenomenon, a facet that remains relatively under-explored within the realm of brain age prediction research using machine learning methods. Voxel-level predictions can provide localized brain age estimates that can provide granular insights into the regional aging processes. This is essential to understand the differences in aging trajectories in healthy versus diseased subjects. In this work, a deep learning- based multitask model is proposed for voxel-level brain age prediction. The proposed model outperforms the model existing in the literature and yields valuable clinical insights when applied to both healthy and diseased populations. Most findings from the analysis align with existing studies on aging, whereas other findings are intriguing and could be potential biomarkers of early-stage neurodegeneration detection. Regional analysis is performed on the voxel-level brain age predictions to understand aging trajectories of known anatomical regions in the brain and show that there exist disparities in regional aging trajectories of healthy subjects compared to ones with underlying neurological disorders such as dementia and more specifically, Alzheimer’s disease. A comparative analysis with traditional deep learning interpretability methods showed that the proposed voxel-level approach to brain age prediction is an effective way to understand regional aging trajectories while being quantitative in nature. The source code is publicly available at https://github.com/nehagianchandani/Voxel-level-brain-age-prediction.
  • Thumbnail Image
    ItemOpen Access
    A working taxonomy for describing the sensory differences of autism
    (2023-04-11) He, Jason L.; Williams, Zachary J.; Harris, Ashley; Powell, Helen; Schaaf, Roseann; Tavassoli, Teresa; Puts, Nicolaas A. J.
    Abstract Background Individuals on the autism spectrum have been long described to process sensory information differently than neurotypical individuals. While much effort has been leveraged towards characterizing and investigating the neurobiology underlying the sensory differences of autism, there has been a notable lack of consistency in the terms being used to describe the nature of those differences. Main body We argue that inconsistent and interchangeable terminology-use when describing the sensory differences of autism has become problematic beyond mere pedantry and inconvenience. We begin by highlighting popular terms that are currently being used to describe the sensory differences of autism (e.g. “sensitivity”, “reactivity” and “responsivity”) and discuss why poor nomenclature may hamper efforts towards understanding the aetiology of sensory differences in autism. We then provide a solution to poor terminology-use by proposing a hierarchical taxonomy for describing and referring to various sensory features. Conclusion Inconsistent terminology-use when describing the sensory features of autism has stifled discussion and scientific understanding of the sensory differences of autism. The hierarchical taxonomy proposed was developed to help resolve lack of clarity when discussing the sensory differences of autism and to place future research targets at appropriate levels of analysis.
  • Thumbnail Image
    ItemOpen Access
    An Investigation of Deep Learning Methods to Shorten GABA-edited Magnetic Resonance Spectroscopy Scan Times
    (2023-09-18) Pommot Berto, Rodrigo; Medeiros de Souza, Roberto; Harris, Ashley; Lebel, Marc; Zhang, Yunyan; Frayne, Richard; McCreary, Cheryl
    Edited magnetic resonance spectroscopy (MRS) can provide localized information on gamma-aminobutyric acid (GABA) concentration in vivo. However, GABA-edited MRS data has a low spectral quality, and many measurements, known as transients, need to be collected and averaged to obtain a high-quality spectrum, resulting in long scan times. This work investigated using deep learning (DL) with only a quarter of the number of conventionally acquired transients to shorten scan times by four while maintaining or improving spectral quality. A proof of concept was demonstrated by reconstructing GABA-edited spectra with only 80 transients and different configurations of DL-based pipelines. The best-performing pipeline used a proposed dimension-reducing 2D U-NET variation and it obtained better spectral quality metrics than conventionally reconstructed spectra with 320 transients. Simulated data was also shown to be useful in pre-training DL model weights. An open data challenge for reconstructing GABA-edited spectra from reduced transients was organized, and various DL models from different participating teams were compared. The challenge results reinforced the proof of concept conclusions that higher spectral quality can be achieved with DL reconstructions. However, the challenge metric evaluation also showed that DL models are able to undesirably exploit the limitations of conventional MRS metrics when using those as the training loss for the models, leading to good metric values but poor reconstructed spectra quality. DL reconstructions of GABA-edited MRS with 80 transients were also quantified and had significant differences from results from conventional reconstructions with 320 transients. However, given the lack of ground truths in the quantified data, it is not possible to conclude which results are closer to the actual concentrations. This work showed that DL methods can reduce GABA-edited MRS scan times while increasing spectral quality. Due to the lack of ground truths for the in vivo data, further studies are necessary to validate the concentrations obtained from the DL-based GABA-edited MRS reconstructions in comparison to conventional methods. This work was developed in the spirit of open science, and the data and code to reproduce the results were made publicly available.
  • Thumbnail Image
    ItemOpen Access
    Brain Magnetic Resonance Spectroscopy: Advances and Applications to Chronic Pain in Knee Osteoarthritis
    (2024-06-24) Leech, Samantha; Manske, Sarah; Harris, Ashley; Dunn, Jeffrey; Ng, Richard; Goodyear, Bradley; Dydak, Ulrike
    This dissertation investigates advancements in brain proton magnetic resonance spectroscopy (1H-MRS) measures and their application to chronic pain in knee osteoarthritis. 1H-MRS measures proton signals, which can be converted into absolute concentrations using the properties of water, brain tissue, and neurochemicals. These concentrations serve as markers of brain health or dysfunction. Current methods to quantify absolute neurochemical concentrations assume an equal distribution of neurochemicals between white matter (WM) and gray matter (GM), an assumption not thoroughly examined. To address this, I determined the distribution of six neurochemicals between WM and GM to establish correction factors to replace assumptions with calculated values. After validation using an independent dataset, I created an open-source tool to implement the calculated correction factors, improving 1H-MRS accuracy by 30-55%. I used quantitative synthetic imaging to measure water properties — relaxation rates (T1 and T2) and proton density (PD) — in different brain tissues of healthy adults. I assessed the impact of inter-individual differences in T1, T2, and PD on neurochemical concentration measures by comparing concentrations calculated using literature-based constants (as is typically performed) to concentrations calculated using individual measures from quantitative maps. In a young, healthy population, individual measures contributed to subtle yet significant variations in calculated neurochemical concentrations, suggesting that using uniform literature values may not be accurate for every individual. Sensitivity analyses indicated that these inaccuracies are likely greater across a wider age range or in individuals with clinical disorders. Applying 1H-MRS, I identified potential neurochemicals and brain regions associated with chronic pain in knee osteoarthritis to understand the brain’s role in this condition. Knee osteoarthritis is a leading cause of chronic pain, with limited research on the specific neurochemicals and brain regions involved. I compared neurochemical levels and their association with pain measures in four brain regions between patients with knee osteoarthritis and healthy controls as well as longitudinally in patients three months after total knee replacement surgery. I found opposing relationships in brain regions associated with pain's sensory and affective dimensions. This dissertation enhances the accuracy of neurochemical concentration quantification and refines the understanding of the brain's contribution to knee osteoarthritis pain.
  • Thumbnail Image
    ItemOpen Access
    Examining Brain Structure after Pediatric Mild Traumatic Brain Injury
    (2022-11-28) Shukla, Ayushi; Lebel, Catherine; Yeates, Keith O.; Harris, Ashley; Brooks, Brian
    Mild traumatic brain injuries (mTBIs) affect millions of children annually and present a huge burden to the public health care system. mTBIs often lead to emotional, cognitive, and physical difficulties, together known as post concussive symptoms (PCS), which usually resolve within 4 weeks of injury. In up to one third of all mTBI cases, PCS can be longer lasting and are referred to as persistent PCS (PPCS). In the pediatric population, since mTBI occurs when the brain is still developing, it can lead to altered developmental trajectories, and consequently affect children's cognitive functioning, symptomatology, and quality of life. This thesis aimed to use advanced neuroimaging techniques, [i.e., diffusion tensor imaging (DTI), neurite orientation dispersion and density imaging (NODDI), and voxel-based morphometry (VBM)] to study unexamined aspects of brain structure associated with mTBI, PPCS, and neurocognitive outcomes after mTBI. I used DTI and NODDI to examine white matter microstructure after mTBI at different time points after injury in comparison to orthopedic injury (OI) and used VBM to examine cerebellar gray matter volume and its association to neurocognitive outcomes of mTBI. The results revealed: 1) No post-acute differences in brain structure (white matter microstructure or gray matter) between children with mTBI or OI, 2) Age moderated differential trajectories of white matter change, 3- and 6-months post-injury in symptomatic children with mTBI compared to asymptomatic children with mTBI and an OI comparison group, 3) Higher gray matter volume in the motor regions of the cerebellum 3-months after injury in the mTBI compared to the OI group, 4) Disruptions in the association between reaction time and cerebellar volume in children with mTBI. This novel set of studies provides new knowledge about brain structure following pediatric mTBI and has important implications for improving our understanding of neurobiological correlates of pediatric mTBI.
  • Thumbnail Image
    ItemOpen Access
    Exploring Region-Specific Changes in Brain Glutamate and Gamma-Aminobutyric Acid (GABA) Across the Migraine Cycle in Children and Adolescents
    (2023-06) Cho, Lydia Youngju; Harris, Ashley; Orr, Serena; Bray, Signe; Jacobs-Levan, Julia
    Migraine is a neurological disorder with multiple phases (i.e., migraine cycle). There is evidence that migraine is associated with excitation-inhibition imbalances and dysregulation of glutamate and gamma-aminobutyric acid (GABA), the primary excitatory and inhibitory neurochemicals, respectively. Magnetic resonance spectroscopy (MRS) is a non-invasive imaging method to quantify brain levels of glutamate and GABA. It has been proposed progressing through the migraine cycle is also related to changes in excitation and inhibition. Few studies have measured glutamate and GABA in adults with migraine, and migraine studies in the pediatric population specifically focusing on glutamate and GABA are very scarce. In this study, children and adolescents with migraine were recruited to have four MRS scans over two weeks to quantify levels of glutamate and GABA in the sensorimotor cortex, thalamus, and occipital cortex. Two approaches were used to determine glutamate and GABA changes across the migraine cycle: (1) binning by migraine cycle phases and (2) time as a continuous metric leading up to or following headache. Analysis of migraine cycle phases showed occipital cortex glutamate was higher in the headache phase and thalamic glutamate was higher in the postdrome phase, both compared to the interictal phase. When using a continuous time metric, glutamate significantly decreased following the onset of the headache phase in the occipital cortex and thalamus, and sensorimotor GABA significantly increased leading up to the headache. I propose that these changes reflect increased excitation in the occipital cortex and thalamus and increased inhibition in the sensorimotor cortex during the migraine attack which may be evidence of thalamocortical dysrhythmia underlying migraine pathophysiology. This study provides insight into the underlying biology of migraine in children and adolescents, and if replicated, may help inform development of future treatments and interventions.
  • Thumbnail Image
    ItemOpen Access
    Exploring the Pathophysiology of Persistent Post-Concussive Symptoms and Metabolite Response to an Aerobic Exercise Treatment Intervention
    (2022-01-17) Joyce, Julie; Harris, Ashley; Yeates, Keith; Debert, Chantel; Lebel, Catherine
    Persistent post-concussive symptoms (PPCS) endure beyond the usual recovery period after concussion. Although the existence of PPCS is well-documented, its pathophysiology is poorly characterized. Emerging research suggests that aerobic exercise may be a viable intervention for PPCS. However, the biological changes that occur with recovery are not well understood. Neurometabolite alterations have been reported widely in mild traumatic brain injury (mTBI) but are rarely investigated in PPCS specifically. Despite their biological relevance to mTBI, the neurometabolites glutamate, γ-aminobutyric acid (GABA) (primary excitatory and inhibitory neurotransmitters) and glutathione (most abundant antioxidant in the brain) are seldom studied. The scarcity of literature on glutamate, GABA and glutathione in clinical mTBI research is likely due to the technical complexity of measuring these neurometabolites non-invasively. This project used advanced magnetic resonance spectroscopy (MRS) to examine glutamate, GABA and glutathione in PPCS patients and age- and sex-matched controls in the anterior cingulate and sensorimotor cortex. Additionally, changes in these neurometabolites were examined in PPCS patients who completed a sub-symptom threshold aerobic exercise intervention. Compared to controls, glutamate exhibited regional reductions in PPCS. Higher GABA was related to greater incidence of previous mTBI. Furthermore, glutathione was related to symptoms of sleepiness and headache burden. The findings of our aerobic exercise intervention for PPCS indicate that exercise may restore healthy cortical metabolism. While these are early results, they provide compelling evidence for the roles of glutamate, GABA and glutathione in PPCS and during recovery with exercise which can inform best practices and the development of other treatments.
  • Thumbnail Image
    ItemOpen Access
    Magnetic resonance spectroscopy in Pediatric Concussion: Examining longitudinal and multisite data
    (2023-10-19) La, Parker Lanhuy; Harris, Ashley; Yeates, Keith; Debert, Chantel; Brooks, Brian
    Millions of concussions occur annually in children and youths. Though metabolic disturbance has been shown to be present in adult concussion, there is less research and evidence of these same effects in pediatric groups. Using Magnetic Resonance Spectroscopy (MRS), we can study neurometabolites that may be altered following pediatric concussion. With increasing interest in MRS to study concussion and other clinical conditions, there is increased importance to understand differences between analysis software and how to harmonize multi-site/vendor studies to increase sample sizes. My thesis aims to improve our understanding of the metabolite changes in pediatric concussion and associated technical moderators in multisite MRS studies. To accomplish this goal, my work examines MRS data from the Advancing Concussion Assessment in Pediatrics (A-CAP) study, which recruited pediatric concussion participants and an orthopedic injury (OI) control group. The results found: 1) No differences in neurometabolites at 12-days post-injury in children with concussion and OI, 2) No significant differences in metabolite trajectories between those with and without concussion symptoms over time (12-days, 3-month, 6-month), 3) Significant decreases in tCho in those with somatic symptoms in comparison to OI controls at 3-months, 4) Large vendor effects and different statistical methods to account for multisite data could yield different results, and 5) Software-related differences in metabolite quantification with associated bias, though clinical results were consistent. These investigations provide important context in pediatric concussion and neurometabolic activity in the short and long-term. Additionally, this work provides recommendations for how to control for multi-site/vendor and offer a description of the biases and agreement between different MRS quantification software.
  • Thumbnail Image
    ItemOpen Access
    Neuroimaging Correlates of Gait Control in Cerebral Amyloid Angiopathy
    (2023-01-16) Sharma, Breni; Smith, Eric E.; Harris, Ashley; Ismail, Zahinoor; McCreary, Cheryl R.
    Cerebral amyloid angiopathy (CAA) is the second most common subtype of cerebral small vessel disease (CSVD) and is characterized by the buildup of beta-amyloid protein in the walls of small-medium sized arteries and arterioles of the leptomeninges. Much is known about common clinical manifestations of the disease, such as presence of white matter hyperintensities, lacunar infarcts, cerebral microbleeds, cortical superficial siderosis, and phenotypic presentations, such as the cognitive profile of CAA and its contribution to neurodegeneration and dementia. However, little had been known about the gait profile of CAA or the neural correlates underlying any abnormalities observed, such as grey matter atrophy, white matter damage, or brain iron accumulation. To address these gaps in the literature, I first conducted a systematic review and meta-analysis of the existing literature covering CSVD and its relation to gait and falls. Once evident that gait difficulties were a feature of CSVD as a whole, I examined gait abilities of patients with CAA, when compared to normal controls (NC), patients with Alzheimer’s disease (AD), and mild cognitive impairment (MCI). With this, I also looked at associations with falls history and fear of falling, as well as the relationships between gait ability and cognition, WMH volume, and CMB count. Significant gait impairments were found in CAA, prompting an examination of associations between these impairments and grey and white matter damage and iron content in select brain regions. In CSVD, there was a general consensus across studies of an association between greater CSVD burden and worse gait and greater falls. Looking specifically at CAA, significant impairments were found in gait compared to NC but not to AD. Further investigation of this lead to associations between worse gait and grey matter atrophy in frontal, AD-affected, and subcortical regions and with greater white matter structural damage. Iron content, however, did not differ between CAA and NC. Overall, gait appears to be negatively impacted by CAA pathology. Further examination of neural correlates may help to better understand the disease and incorporation of the current findings may help to inform clinicians on the functional outcomes of CAA patients.