Open Theses and Dissertations
Permanent URI for this collection
This collection is the result of a joint project between the Faculty of Graduate Studies and Libraries and Cultural Resources which provides Graduate students with the opportunity to archive their thesis with University Archives in our digital repository.
If you are a Graduate student submitting your final thesis to PRISM, please ensure you have read and submitted all required documents: http://grad.ucalgary.ca/current/thesis
If you require assistance submitting your thesis please contact thesis@ucalgary.ca
The electronic theses and dissertations on this site are for the personal use of students, scholars and the public. Any commercial use, publication or lending of them in libraries is strictly prohibited.
Browse
Browsing Open Theses and Dissertations by Author "Abbasi, Zahra"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
- ItemOpen AccessExploration of Techniques for Working with Sparse Data when Applying Natural Language Processing to Assist a Qualitative Data Analysis of a COVID-19 Open Innovation Community(2024-04-17) Yamani, Shirin; Barcomb, Ann; Far, Behrouz; Abbasi, ZahraThis thesis undertakes a novel integration of Natural Language Processing (NLP) with Qualitative Data Analysis (QDA) to investigate the dynamics of volunteer involvement within the TeamOSV community, a collective formed in response to the COVID-19 pandemic. Central to this study is the exploration of roles and interaction patterns among episodic and habitual volunteers, alongside an analysis of the factors influencing their engagement and disengagement within the community. A significant methodological contribution of this work lies in addressing the sparse data challenge, a common constraint in qualitative research, particularly within multi-class classification contexts. The study employs and critically evaluates a range of NLP techniques, with a focus on data augmentation strategies, to enhance the efficacy of various models, including Logistic Regression, Naive Bayes, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), and particularly the Self-Attention model. The proposed framework, identified for its superior performance, demonstrates a noteworthy ability to process and interpret sparse qualitative data, surpassing both traditional approaches in its effectiveness. Furthermore, the thesis explores an in-depth analysis of model variations, assessing the impact of differing configurations of Self-Attention blocks and layers of feed-forward neural networks. It also explores the implications of pre-training on model performance, offering insights into the architectural complexities and training dynamics of NLP models. A crucial aspect of this exploration is the consideration of the trade-offs between model complexity and computational efficiency, highlighting the practical challenges and considerations in deploying these models in qualitative research contexts. Qualitatively, the study offers a detailed examination of volunteer roles within the TeamOSV community. It identifies the distinct contributions and challenges associated with episodic volunteers, characterized by their sporadic engagement patterns, and habitual volunteers, who provide stability and long-termvision. The research also sheds light on the reasons behind volunteer disengagement, such as lifestyle changes and diminishing interest, providing a holistic understanding of volunteer participation in open-source, community-driven projects. The thesis concludes by emphasizing the collaborative strengths of merging NLP with QDA, a union that significantly augments the depth of qualitative research. It proposes a roadmap for future investigations, concentrating on enhancing insights into volunteer coordination within open innovation settings and broadening the application range of NLP in qualitative data examination.
- ItemOpen AccessImproved Cavitation Monitoring and Detection Methods for Focused Ultrasound Blood-Brain Barrier Disruption(2022-01) Khan, Sonia; Curiel, Laura; Curiel, Laura; Smith, Michael; Fear, Elise; Abbasi, ZahraTranscranial focused ultrasound (FUS) in combination with microbubbles has demonstrated promising outcomes in treatment of brain disorders by stimulating transient BBB disruption, allowing therapeutics to enter into the brain. Currently, time signals from a hydrophone are transformed into frequency domain to monitor cavitation activity during BBB opening. The area under the curve (AUC) in a 300 Hz bandwidth around the subharmonic is used as a metric to determine cavitation activity. However, given the available frequency resolution, there are very few points within the 300 Hz bandwidth for precisely analysing the AUC. Also, many recorded signals show no detectable subharmonic above the noise as the A/D calibration is overwhelmed due to the strong fundamental. These issues can result in the subject being under treated or over treated. This research aims to better monitor and control the cavitation phenomenon for BBB disruption by developing methods to improve the cavitation spectra. The acoustic signals captured by a hydrophone from the excited microbubble phantom were filtered with a low pass analog filter to improve the system’s dynamics of self-calibration by suppressing the strong fundamental. The acoustic signals were further improved with two proposed signal processing techniques, namely, Fourier interpolation via zero-padding to increase the spectral frequency resolution, and windowing, which allowed us to uncover previously unreported subharmonic side lobes. Additionally, we propose the bandwidth to be wider than the current 300 Hz for AUC to include the useful information in these side bands. Our proposed improvements were validated on animal data. Finally, the performance of our proposed improvements was compared to traditional methods by evaluating metrics for cavitation detection. Previous studies reported a steady increase in the AUC with increase in pressure, whereas our work presented that the AUC would rise drastically at the stable cavitation threshold indicating that maximum energy is concentrated in stable cavitation regime. Then, beyond this threshold, the AUC will drop before rising again, signifying the energy shift towards initiating inertial cavitation. Our findings can be beneficial to enhance cavitation detection metrics and we can actually visualize the different cavitation regimes when the AUC and power are evaluated for subharmonic.
- ItemOpen AccessPedersen Conductance as Measured by Swarm(2023-07) Pourkarim, Pouya; Knudsen, David; Burchill, Johnathan; Cully, Christopher; Abbasi, ZahraWe estimate ionospheric Pedersen conductance using measurements from the European Space Agency’s Swarm A satellites of magnetic and electric fields. We provide long-term averages in the form of global maps, and case-by-case results using scatter plots of Pedersen conductance versus solar zenith angle. The long-term results display expected ionospheric features and show agreement with previous studies. However, we find that the case-by-case results display a high variability, and abnormally low Pedersen conductance values. We identify possible sources of the high variability and low magnitude, and we revisit the underlying assumptions conventionally used in such calculations. We find that variation of ionospheric parameters in the zonal direction, and contamination of zonal magnetic field by Hall currents are potentially significantly affecting our conductance estimates. Further, we identify non-quasi-static fields and presence of Alfvén waves as another significant source potentially affecting the results.
- ItemOpen AccessSelf-healing of Direct Written Conductive Inks for Curvilinear Circuits(2023-04-18) Jeong, Chan Woo (Robin); Park, Simon; Du, Ke; Abbasi, Zahra; Komeili, AminThe increased electrification of vehicles in both automotive and aerospace industries has introduced new challenges in manufacturing complexities and weight management. Complex and heavy wirings are currently being utilized and conventional printed circuit board (PCB) manufacturing methods are limited in 2D geometries. Alternatively, a direct-writing approach presents weight and materials saving opportunities where planar substrates with circuits already printed are formed to a final shape. However, designing a circuit or a printed ink formula able to withstand the high strain of substrate forming is challenging. Instead, a circuit able to regain functionality after sustaining strain induced cracks presents a more versatile approach. In this study, a conductive ink with self-healing capabilities is developed. A copper-nanoparticle based ink compatible with existing lithographic methods is developed and printed on planar polymeric substrates. Intense pulsed light (IPL) is utilized to photothermally heat, reduce, and sinter copper nanoparticles within milliseconds. By utilizing light-matter energy absorption and the plasmonic effect, heat sensitive polymeric substrates are unaffected while conductive copper tracks are formed. After printing, drying, and IPL flashing, the substrate and printed tracks are subjected to cyclic bending and thermoforming. Afterwards flashing is performed once again to initiate the healing process through reflow of indium microparticles. These indium healing agents added to the ink bridges microcracks via capillary forces to recover severed electron pathways. Mechanisms of photothermal heating and sintering is simulated to better understand the underlying physical phenomena. Ultimately, a planarly written copper nanoparticle ink capable of surviving substrate deformation to produce curvilinear circuits is achieved. This direct writing method can provide drastic wiring weight reduction imparting fuel savings in the next generation of electronics in vehicles.
- ItemOpen AccessSensitivity-enhanced Microwave Sensors for Real-time Detection and Monitoring(2024-05-27) Vestrum, Sarah Viola; Abbasi, Zahra; Abbasi, Zahra; Murari, Kartikeya; Badv, MaryamPlanar microwave sensors have gained popularity due to their real-time, non-invasive sensing abilities. These structures have successfully enabled various range of applications in various applications, from small-volume liquid characterization in biomedical applications to sensing and detection in high-pressure and temperature environments. While planar resonator structures were introduced to the filter design domain first, they have transited into an ideal candidate for real-time sensing and monitoring to address different limitations that waveguide microwave sensing approaches suffer from, including bulky structures and requiring higher volumes of the sample under the test. This work focuses on enhancing the sensitivity of microwave structures using single-port reader-tag based structures. Unlike the popular two-port planar microwave sensor structures, single-port structure designs have the advantage of lowering the requirements and costs for measuring equipment, making them suitable for personalized sensing applications. Here, three single-port reader-tag based planar sensors have been introduced to enhance sensitivity and sensing distance for different rapid liquid characterization applications in the medical field. First, a patch antenna sensor for distant, small volume water-content detection. This structure detects water absorption with a resolution of 25 μL using a hydrogel-integrated sensing tag to improve sensitivity. Then, a patch antenna sensor for distant electrolyte concentration detection in urine for hydration monitoring was developed. The fabricated sensor was able to detect concentration changes of 0.5% at a distance of 24 mm from the reader, making it a well-fit candidate for wearable dehydration monitoring applications in older adults due to their increased susceptibility to dehydration.