14277 research outputs found
Sort by
Development and applications of a fault-tolerant neuroscience platform: from understanding healthy brain development to Alzheimer’s disease progression
No full textMany neuroscientific studies transform Magnetic Resonance Imaging (MRI)
data using an intricate cadence of data processing, extraction, and featureengineering
steps to derive the datasets necessary to accomplish a given
research objective. These processing steps may require multiple software
systems across different software modalities leading to a fragile and difficultto-
replicate task-specific pipeline. Thus, there is a need in neuroscience research
for a pipeline management platform that is extensible, vendor-neutral,
facilitates interdisciplinary collaboration, supports reproducible research, and
can simplify the complex orchestration of measurement extraction from raw
brain data. In this study, a pipeline orchestration software was developed and
its use and extensibility were demonstrated across three novel and distinct
neuroscience-related research studies. The first study introduces our software
in the context of its ability to execute pipelines on a large dataset and
produce insights expected of neurotypical participants. The second research
study identifies novel biomarkers of early-stage Alzheimer’s disease and compares
their predictive performance against known biomarkers. The third study
investigates the typical neural development of cognitive abilities in schoolaged
children, focusing on language acquisition, and examines potential associations
with environmental and developmental factors which support the
theory of interactive specialization. Our three studies demonstrate the wide
applicability of our pipeline orchestration software to address neuroscience
research questions, and its support to reproducible research. Our software
(AirCRUSH) is available at https://github.com/dmattie/aircrush,
https://github.com/dmattie/aircrush-core-operatorsIncludes bibliographical references (pages 151-177
From local to regional seabed maps: Developing the methods and data products needed for marine conservation planning in Newfoundland and Labrador
No full textMaps are essential tools for understanding the environment around us. This is especially true for the global ocean: the majority seafloor habitats are out of sight and beyond the reach of conventional survey methods. In this thesis, I use two case studies to demonstrate the utility of multibeam echo-sounding and biological sampling to map marine habitats in support of conservation planning. Habitat protection is a key pillar in the conservation strategy outlined by Canada’s Species At Risk Act, yet information on critical habitat is not available for most marine species at risk. In Chapter 2, I use high-resolution multibeam, physical samples, and video surveys to identify biologically distinct habitats and delineate potential spawning and nursery areas for Atlantic wolffish. In addition to characterizing and mapping seafloor habitats, this work highlights potential vulnerability of nearshore wolffish spawning habitats to warming coastal waters. Habitat mapping is also important to Marine Protected Area (MPA) design and monitoring. In Chapter 3, I define and map the seafloor habitats of Newfoundland and Labrador’s Eastport MPA to assess MPA design against stated management goals. Despite aiming to protect endangered wolffish, this small MPA does not include suitable wolffish habitats and, unfortunately, contributes little to regional biodiversity. This work highlights the importance of science-driven management and the challenges faced when single-species fisheries closures are redefined as broader conservation measures without adaptive management to support the expanded objectives.
These case studies add to mounting evidence that further investment in and effective use of marine habitat maps is key to effective conservation and sustainable management of our oceans. However, multibeam surveys are expensive and time-consuming. In Chapter 4, I outline a method for using low-cost crowd-sourced data to improve seafloor maps at a regional scale and
higher spatial resolution than previously possible. I use this method to produce novel data products for over 670,000 square kilometres of the Newfoundland and Labrador shelf, including identification of almost 2000 km2 of previously unmapped tributary submarine canyons. This research advances our understanding of Newfoundland and Labrador waters and presents tools that can be applied to science-based marine conservation planning regionally and globally.Includes bibliographical references (pages 111-138
Exploring motor imagery–induced neural activation and corticospinal excitability in healthy adults using EEG and TMS
No full textMotor imagery (MI)–based brain–computer interfaces (BCIs) activate motor–related brain regions and, through neurofeedback, foster neuroplasticity, offering significant potential for neurorehabilitation [1]. While MI–BCIs have shown success in restoring hand function after stroke, their use in mitigating upper–limb impairments in multiple sclerosis (MS)—a chronic neurodegenerative disorder characterized by impaired motor control and coordination—remains underexplored [2]. This study presents a preliminary investigation into the feasibility of using MI–BCI for targeted therapy of two specific hand motor function deficits common in MS: fine motor control, represented by a task involving repeated cycles of full hand extension and thumb–fingertip closure, and motor coordination, represented by a sequential finger–to–thumb opposition task. Using two hand tasks representing each of these functions, the objectives were: 1) to examine whether MI of these tasks alters corticospinal excitability (CSE), indicating the potential for fostering neuroplasticity and improving hand function, and 2) to determine if distinct task–specific neural activation patterns can be detected via electroencephalography (EEG), suggesting they may be targeted with BCI–based neurofeedback training.
Data from twenty–one healthy participants (8 males, 13 females; mean age: 41.35 ± 8.36 years) were analyzed for this study. Transcranial magnetic stimulation (TMS) was used to assess changes in CSE due to MI of the coordination and control tasks. A single TMS pulse was delivered during intervals of MI and rest and resulting motor–evoked potentials (MEPs) were measured from the first dorsal interosseous (FDI) muscle. For the control task, MI did not significantly increase MEP amplitudes compared to rest (Δ=39.91 μV, p=.354), but significantly decreased MEP latencies (Δ=−0.34 ms, p=.002). For the coordination task, MI also did not significantly increase MEP amplitudes (Δ= 29.67 μV, p = 0.243), but significantly decreased MEP latencies compared to rest (Δ=−0.30 ms, p = 0.006). CSE was also assessed during actual execution of the tasks (ME), and changes in both MEP amplitudes and latencies were significantly different from rest for both tasks (control: Δ=515.96 μV, p < .001; Δ=−1.27 ms, p < .001; coordination: Δ=534.98 μV, p < .001; Δ=−1.15 ms, p < .001).
In a separate session, 64–channel EEG was recorded as participants performed 60 intervals each of MI and ME of the two tasks, as well as rest. To determine if the control and coordination tasks were distinguishable using EEG, several different classification pipelines were explored. In terms of feature extraction, Filter Bank Common Spatial Patterns (FBCSP) based on the five standard EEG frequency bands (delta, theta, alpha, beta, gamma), as well as on a finer set of nine overlapping bands spanning 1–40 Hz in 4 Hz increments, was explored, as were functional connectivity–based features, specifically correlation, coherence, and phase–locking value (PLV). Classification was performed using multiple algorithms, including Linear Discriminant Analysis (LDA), Support Vector Machines (SVM), and Decision Trees (DT). Using the PLV features, classification of control vs. coordination tasks yielded high accuracy (80.5% for MI and 82.7% for ME using a SVM classifier). Accuracies above the chance threshold (i.e., 58.3% for n=120 trials, α=0.05, based on binomial distribution) were obtained for all subjects for both MI and ME, and a majority of participants achieved accuracies above 70%—19 participants for MI and 17 for ME— indicating individual potential for task–specific decoding.
Although MI did not significantly increase MEP amplitude, the consistent latency reductions suggest enhanced corticospinal conduction, likely reflecting subthreshold motor activation. These results may support the potential of MI to facilitate neuroplasticity, despite limited task–specific modulation in excitability. On average, the distinct neural patterns differentiating motor control and coordination were detected with accuracy greater than chance, and for some participants with very high accuracy. Together the EEG and TMS results support the feasibility of targeted BCI–based therapy for improving motor control and coordination. Further work is needed to more reliably and specifically identify the neural activation patterns associated with these functions, particularly in people with MS, and then to evaluate the efficacy of BCI–mediated MI therapy for improving upper limb function in the target population
Modelling the impacts of projected climate-driven changes in return period rainfall on peak flows of the Upper Humber River, Newfoundland
No full textAs climate change threatens the resilience and well-being of rural communities across Canada, this study seeks to fill gaps in knowledge by investigating the impacts of potential future hydrological intensification on peak flows of the Upper Humber River in rural Newfoundland. A basin model of the Humber River watershed was generated using the Hydrologic Engineering Center's Hydrological Modeling Software (HEC-HMS), utilizing available Digital Elevation Model (DEM), soil and land use data. Observed rainfall and streamflow data were used to calibrate and validate the model, which was found to simulate streamflow with statistical accuracy. Return period peak flows under current climate conditions were estimated using historical rainfall data, producing results consistent with previous studies. Estimated return period peak flows for the 2020s, 2050s, and 2080s utilized rainfall projections considering an intermediate greenhouse gas (GHG) emission scenario (RCP 4.5). This study projected a rise in the frequency, magnitude, and uncertainty of peak flows with the projected progression of climate change, with the most pronounced changes occurring later in the century. At Humber River inlet into Deer Lake, a 40% increase in the magnitude of median 100-year return period peak flows was projected from the baseline historical period (1966-2014) to the 2080s, emphasizing the urgent need for comprehensive planning, continued climate change research, and the implementation of targeted flood adaptation strategies within the Humber River watershed to reduce the potentially catastrophic impacts of projected future climate change
The hidden costs of rigidity: how class-based data limits machine learning performance, increases omitted variable bias, and reduces repurposability in crowdsourcing
No full textThe design of data collection interfaces can significantly influence the quality of training data for machine learning (ML) systems. The predominant way designers collect data is class-based, which results in the failure to capture potentially useful data. This thesis proposes an alternative data instance-based data collection method, which retains as much data as possible. This thesis compares the rigidity of class-based data collection to the more freeform instance-based data collection, and assesses the impact of each approach on ML performance. Evaluation is conducted using a controlled experiment using autonomous driving scenarios, a domain where data completeness is critical for safety. Participants (N=257) classified driving scenes via one of two interfaces: class-based (fixed categories) and instance-based (open-ended descriptions). Machine learning models were trained on datasets derived from both methods and compared for predictive accuracy. Results show that instance-based data substantially outperformed class-based data across all models, with dramatic improvements in accuracy and robustness. The instance-based approach increased the number of granular and diverse attributes captured, which directly enhanced ML performance by enabling models to learn more comprehensive patterns. Furthermore, the richer and more granular data collected through open-ended reporting demonstrated greater adaptability for alternative analytical tasks, supporting applications beyond the original scope of data collection. The flexibility of instance-based data collection mitigated omitted variable bias by capturing contextual attributes often excluded in class-based design, thereby enhancing the models' ability to generalize to real-world complexity. These findings highlight the importance of representational flexibility in data collection designs, challenging the dominance of class-based approaches. In cases where data is crowdsourced to train machine learning models, this work demonstrates the potential benefits of instance-based data collection design, which integrates structured input with open-ended reporting to enhance ML performance
Enriched housing alters maternal response and promotes sex differences in offspring development in house mice (Mus musculus domesticus)
No full textLaboratory mice are born with immature nervous systems and require constant care for
development. Pups use ultrasonic vocalizations (USVs) to communicate with their mother (dam).
Preclinical rodent models of neurodevelopmental disorders (NDDs) use USVs to measure social
development, but maternal-offspring dynamics are overlooked. This study investigates whether
environmental differences influence such dynamics in a sex-dependent manner by combining
behavioural testing and quantitative polymerase chain reaction (qPCR) on brain tissue. Female
C57BL/6 mice were assigned to standard (ST) or enriched (EE) housing and mated. EE housing
was larger, with extra bedding and enrichment items. Litters were culled to four pups (2/sex);
USVs and maternal response were recorded on post-natal days (PND) 6 and 8. ST dams were
more pup motivated and attentive to females than EE dams. QPCR data indicate that co-
regulation of genes linked to maternal response and stress reactivity is driving such behaviour.
ST housing suppressed USVs, and male pups showed genetic markers of decreased
neurodevelopment and disruptions in stress response and sexual differentiation. EE dams spent
equal time with the sexes, representing a semi-natural response that altered pup USVs and
markers of neurodevelopment in males. Preclinical NDD work must consider environmental
effects on maternal-offspring dynamics when interpreting developmental sex differences
Seasonal Trends in Water Retention of Atlantic Sea Cucumber (Cucumaria frondosa): A Modeling Approach
No full textSea cucumbers are widely consumed as a delicacy or in eastern medicine across
many Asian countries. Due to the depletion of traditional stocks, new species are increasingly
harvested, including the Atlantic sea cucumber (Cucumaria frondosa), the most
abundant, cold-water species found in the North Atlantic. This species is harvested in
NAFO subdivision 3Ps off the south coast of Newfoundland and Labrador, Canada. As
part of their respiration, stress response, and locomotion, sea cucumbers draw and retain
oxygenated water within their body cavity, resulting in significant water content at
landing. Historically, Fisheries and Oceans Canada (DFO) have applied a 23% deduction
to the landed weight to account for this water retention. To validate this deduction, the
authors conducted experiments across thirteen sampling events in 2019 and 2020. Randomized
samples were collected during offloading and were categorized into three sizes of
bin—small (x ≤ 150 g), medium (150 g < x ≤ 250 g), and large (x > 250 g)—and water loss
was measured. Water loss was analyzed in relation to multiple factors, including processor,
unloading method, year, license, month, fishing area, hold location, size, and processing
method. Key findings included the following: (a) sea cucumbers typically contained more
than 23% free water; (b) large and medium-sized specimens, which dominated landings,
retained more free water; (c) water loss was highest for the samples collected from the top
of the hold; (d) the unloading method influenced free water retention, as did the processing
method used to cut the sea cucumbers; (e) license, processor, and fishing area had strong
collinearity with other factors or were not found to be statistically significant; and (f) water
loss appeared higher in 2020 than 2019, largely due to the increased use of vacuum transfer
methods. Based on these findings, DFO revised the water retention allowance to 34%
Getting Off the Boat: Re‐Considering Research Responsibility and Knowledge Dynamics in Ocean Literacy
No full textIn light of the UN Ocean Decade’s calls for increased ocean literacy, what can critical perspectives on inter‐epistemic exchanges contribute to the practice of researchers themselves? Herein, we aim to expand on scholarship analyzing the relationship between researchers and local/Indigenous knowledge holders beyond notions of knowledge commensurability, towards interpersonal practices. A framework of relationship‐building allows local perspectives and knowledge to be included both actively and passively in research. However, this requires marine scientists to spend time disembarked from sampling vessels in local communities. This adaptation in research methodology involves the scientist becoming a person first, and a researcher second. A paradigm shift occurs where the researcher’s function is that of a guest, whose primary exercise is to actively listen. This repositions ocean literacy as a reciprocal process, whereby the scientist learns from diverse perspectives to inform and enrich mutual understandings of the ocean. We build here on research experiences to show how interpersonal relationships, rather than systemic ones, can help build richer collaboration. This dynamic is illustrated through the case of a marine habitat mapping study in the Canadian Arctic. Community engagement was prioritized by the researcher as a first step, allowing for exposure to local understandings of the ocean to orient research questions. Outcomes included locally relevant marine maps and research findings, culturally responsive outreach materials, a recovered airplane, short‐term local employment, and long‐term relationships which continue to the present day. This case demonstrates how the intentional development of interpersonal relationships can leverage research activities towards building ocean literacy which respects and recognises diverse knowledge systems
Material Communities: Urban Reuse Centres and Community-Oriented Circularity in the Building Sector
No full textThe version available in this research repository is a preprint. Its content does not reflect the peer-review process and it lacks publisher layout and branding.This paper investigates ‘urban reuse centres,’ nonprofit entities promoting sustainable practices in Construction, Renovation, and Demolition (CRD). Across North American cities, these centres address challenges from shifting local policies on demolition and landfill and increasingly serve as hubs for social and environmental initiatives, including the Circular Economy (CE). Despite their importance, academic scrutiny is lacking. This study critically examines reuse centres as integral components of an evolving community circular economy for the CRD sector. Through expert interviews, we demonstrate how urban reuse centres challenge conventional CRD practices, foster novel engagements with circularity, and navigate frictions in promoting reuse within communities
Enzymatic incorporation of docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) into medium chain triacylglycerol oil (MCTO) and virgin coconut oil (VCO): positional distribution and oxidative stability of structured lipid products
No full textThe lipase-catalyzed acidolysis for producing structured lipids from medium-chain triacylglycerol oil (MCTO) and virgin coconut oil (VCO) with long-chain n-3 polyunsaturated fatty acids (n-3 PUFAs), specifically docosahexaenoic acid (DHA) and/or eicosapentaenoic acid (EPA) was investigated. Three commercial enzymes, namely Thermomyces lanuginosus lipase, Rhizomucor miehei lipase, and Candida rugosa lipase were used as biocatalysts. Among them, immobilized Lipozyme® TLIM from Thermomyces lanuginosus demonstrated the highest degree of DHA or EPA or DHA+EPA incorporation into both MCTO and VCO.
The effects of varying reaction parameters, including the mole ratio of substrates, enzyme load, reaction time, and reaction temperature, were monitored to determine the most effective conditions. Incorporation of n-3 PUFAs into MCTO and VCO increased significantly (p<0.05) with increasing mole ratio of substrates. As the enzyme load increased from 1 to 4%, the incorporation of n-3 PUFAs also increased; however, it decreased when the enzyme load was further increased to 6%. The incorporation of these fatty acids increased with reaction time, from 12 to 36 h, but decreased at 48 h. Similarly, the incorporation of n-3 PUFAs increased with temperature from 35 to 45 oC and decreased at 55 and 65 oC. The highest incorporation of n-3 PUFAs was achieved at a mole ratio of 1:3 (MCTO or VCO to DHA or EPA). The maximum incorporation of DHA+EPA occurred at a mole ratio of 1:3:3 (MCTO or VCO to DHA+EPA).
Response surface methodology (RSM) was employed to maximize the incorporation of n-3 PUFAs while minimizing enzyme usage. The process parameters studied included enzyme amount (2, 4, 6%), reaction temperature (35, 45, 55 oC), and reaction time (24, 36, 48 h). All experiments were carried out using a central composite design (CCD). Under optimal conditions of 3.3% Thermomyces lanuginosus enzyme, 42.22 oC, and 33.37 h, the incorporation of DHA was
45.0% in MCTO and 32.9% in VCO. Optimization of acidolysis with EPA resulted in a maximum EPA incorporation of 47.4% in MCTO and 44.5% in VCO. Similarly, the maximum incorporation of DHA+EPA was achieved with 50.4% in MCTO and 47.0% in VCO.
Another study on stereospecific analysis was conducted to identify the positional distribution of fatty acids in the triacylglycerol (TAG) of n-3 PUFAs-enriched oils. In n-3 PUFAs-enriched MCTO, these n-3 PUFAs were predominantly esterified at the sn-1 and sn-3 positions, while C8:0 and C10:0 were mainly located at the sn-2 position. Similarly, in n-3 PUFAs-enriched VCO, these n-3 PUFAs were primarily attached to the sn-1 and sn-3 positions. C12:0 was mainly esterified at the sn-3 position, and C8:0 and C10:0 were distributed randomly across all three positions.
The oxidative stability of enzymatically modified oils, as well as their unmodified counterparts, was evaluated under Schaal oven conditions at 60 oC over a 12-day storage period. The assessment involved measuring conjugated dienes (CD), 2-thiobarbituric acid reactive substances (TBARS), and headspace volatile compounds. Among the oils examined, the enzymatically modified products exhibited higher levels of CD and TBARS compared to their unmodified counterparts. Additionally, the modified oils showed a significantly increased (p<0.05) rate and extent of lipid peroxidation, as indicated by the rising of CD and TBARS values and the accumulation of volatile compounds over the storage period. The primary volatile compounds identified in the enzymatically modified oils included acetaldehyde, propanal, and acrolein, which were not present in the unmodified oils.Includes bibliographical references (pages 195-253