21793 research outputs found
Sort by
Hysteric: Women’s Histories and the Psychoanalytic Aesthetic from Charcot to Rego
Get PDFPaula Rego (b. 1935) is a Portuguese-British artist whose work has been significantly informed by histories of feminism. Her responsive and reactive work often takes direct inspiration from historic events. In her 2004 series of pastels titled Possession I-VII, Rego adapts photographs commissioned by Jean-Martin Charcot of patients at the Salpêtrière Hospital, positioning her model, Lila Nunes, on an analytic couch. In doing so, Rego utilises the aesthetics and history of psychoanalysis in a careful re-adaptation, allowing us to reconsider what it is to depict a subject in this way. This thesis examines the Charcot photographs, as well as the depictions of Dora by Hélène Cixous and Sigmund Freud as means to trace the development and the changing perceptions of “the hysteric” or“madwoman”. By considering the history of this figure, her representations, and the ways she has been treated through the nineteenth to twenty-first centuries, I aim to shed new light on how this history effects the way we think about women’s pain, agency, emotions, subjectivity, sexuality, and ways in which they take up space in our own contemporary society
Domain Adaptation Methods for Sparse Coding Based Non-Intrusive Load Monitoring
Get PDFEnergy disaggregation, or Non-Intrusive Load Monitoring (NILM), is a technique that predicts the consumption levels of individual appliances from only the main signal in the building. Various methods have been proposed to solve this problem, including sparse coding (SC), which offers great advantages due to its ability to capture complex patterns in data. However, a challenging aspect of NILM is that data containing appliance-level information is scarce. Moreover, the houses that the models are tested on might be from a different population than the training data, thus resulting in a domain shift. Therefore, we need to develop approaches that are adapted to training data scarcity through the use of transfer learning (TL), also known as domain adaptation. In this research work, we explore domain adaptation approaches on SC models with the aim of discriminative energy disaggregation (DD). We compare 4 methods that employ TL, two of which are deep architectures, with 4 methods that do not employ it. In the second part of this thesis, we explore constraining NILM domain adaptation to a privacy-preserving Federated Learning framework. In this case, the NILM models are being trained in a framework that does not allow data to be shown to any model outside of the building's domain. This allows us to experiment with distributed methods in a more realistic setting, where user data is omitted from any third party. For this task, we propose 4 weighted federated domain adaptation methods. We also experiment with weighting methods that further protect the privacy of the user, resulting in a total of 12 approaches that we compared
A Rogers-Shephard Type Inequality for Surface Area
Get PDFThe famous Rogers-Shephard inequality states that, for any convex body , we have a volumetric inequality that compares the volume of the difference body of , , with the volume of . Using Cauchy's surface area formula, a particular case of the more general {\it{Kubota's Formulae for Quermassintegrals}}, we extend this classical inequality to extrapolate an upper bound for the surface area of the difference body within the Euclidean space . We accompany this upper bound with a lower bound that we derive from the classical Brunn-Minkowski inequality, . Embracing a geometric perspective, we delve into the nuanced relationships between convex bodies and their respective surface areas, scrutinizing the patterns and properties of the difference body. This includes the validation of the upper bound in for certain classes of convex bodies, including smooth bodies and polytopes. We then analyse and establish a pattern for the formation of the difference body of pyramidal structures in . Finally, we draw a conclusion on the effect of symmetry on 's proximity to either bound. More specifically, we observe how deviations from symmetry, mainly when is a simplex, often considered the most asymmetric convex body, draws the constant closer to the upper bound
An Articulation of Becoming: Transformation and agency through studio inquiry
Get PDFThis thesis employs a research-creation methodology to examine the relationship between an artist and their body as experienced through the multifaceted role of identity. It responds to the role of agency in artmaking through the practice of performative photography while developing a series of works featuring dynamic images of the body in motion. The study examines the history of movement in photography, interpreting aesthetic outcomes and contemplation of performing for the camera. With a feminist/ethnic (Other) lens, the art practices of four performance artists from Cuba, Chile, and Colombia are explored, and their artistic approaches are observed through themes of exile, the political body, the collective body, and ritual. The research is carried out while considering how this author’s work is impacted by her identity and lived experience. The studio inquiry discusses the exhibition titled Impulse: A Photographic Exhibition on The Articulation of Becoming, where this work was presented. As a research-creation, it employs performative photography to analyze the transformative ability of art that connects the self to the collective narratives. Themes of identity and selfhood, as well as notions of the body in art, can be valuable creative access points in the domain of art education, demonstrating that art as pedagogy can bridge knowledge across disciplines, ultimately contributing to a deeper understanding of our socio-cultural contexts
Pigment-protein Complexes of Photosynthetic Bacteria: Convex-Lens Induced Confinement Microscopy and Single Molecule Spectroscopy Simulations
Get PDFThis project involved studying photosynthetic pigment molecules and had two distinct
sections of the project: an experimental part and a computational part.
Throughout the experimental part of this project we explored several different
modifications to our experimental setup to overcome a lack of flatness and repeatability in
our sample. We realized that there may have been non-elastic properties to the gaskets we
were using, and we found that CLIC microscopy was not suitable for our purposes.
During the computational component, we generated a predictive model which more
than doubled the speed of generating appropriate energy landscapes for spectral hole burning
simulations for realistic values of md2
. We learned that an increase in md2
decreases the
spacing of the energy bands, and increasing the standard deviation of the wells increases the
spacing of energy levels within the same band. We developed a Markov chain Monte-Carlo
single molecule spectroscopy algorithm and implemented it into the larger program. We
observed several instances where the pigment molecules can oscillate between two wells very
quickly. This is a result of randomly generating landscapes where sometimes adjacent wells
with similar energy levels and low barriers are generated which creates exceptionally high
tunneling rates.
By comparing our histogram to the LH2 experimental results of Köhler’s group, we
were able to make a prediction of md2 =1.34*10-26 kg nm2
. This matches well with values
obtained for several other complexes
Characterization of ORF19.7608 (PPP1), a Biofilm-induced Gene Encoding a Protein of Unknown Function in Candida albicans
No full textBiofilms are a major source of pathogenicity in Candida albicans and are connected to about half the deaths due to systemic candidiasis. Biofilm formation is controlled by a transcriptional network and involves over a thousand genes including a small gene we have named PPP1 (Punctate Pattern Protein 1). This is a biofilm-upregulated gene encoding an unexamined protein found only in Candida albicans and its sister species Candida dubliniensis. In this study, we characterize this protein by analysing its effect on biofilm formation and its cellular localization. We focus on fluorescent microscopy to identify its subcellular localization and links to other cellular processes. First, we tagged Ppp1 with the green fluorescent protein (GFP) and imaged the protein expression pattern. This expression pattern was used to identify subcellular compartments with similar patterns and direct knockouts and tags of proteins defining these domains. Next, we screened the mutant and complement strain through stress and invasive assays and confocal microscopy to identify a distinct phenotype. Disruptions and tagging of Sur7 defining the eisosome and Arc35 defining the actin cytoskeleton suggested that Ppp1 was not a component of either the eisosome or actin cytoskeleton. However, tagging Sed5, a Golgi defining protein, identified colocalization between the puncta. Since there is no clear distinction between Golgi compartment proteins and secretory proteins, we tagged and analyzed Sap2 and compared data with results for Ppp1. We also tagged Orf19.6274 and Orf19.4654, proteins with similar characteristics to Ppp1 and compared their patterns to Ppp1 puncta. These experiments support a model that Ppp1 is localized in the Golgi
Exploring Natural Product Biosynthesis in Photorhabdus laumondii: A Novel Strategy through Activation of Bacterial Enhancer Binding Proteins.
Get PDFIt has been reported that the number of known natural products is much less than the number of biosynthetic gene clusters (BGCs) found in the genomes of microorganisms, which suggests a considerable potential for new drug discovery. However, a significant challenge arises as these BGCs cannot be induced in a laboratory setting. Due to high fitness cost for bacteria, these encoded natural products are tightly regulated. Consequently, our limited understanding of the triggers and regulatory mechanisms of these BGCs impedes progress in the search for potential antibiotics.
To tackle this issue, we propose an alternative approach distinct from the traditional methods such as high-throughput screening, heterologous expression, or the introduction of constitutive promoters. Our strategy involves using the bacterial machinery to activate the expression of BGCs by constitutively activating their bacterial enhancer binding proteins (bEBPs). This is achieved by deleting the regulatory domain located at the N-terminal of bEBPs. The majority of these bEBPs are responsible for activating the σ54-RNA polymerase holoenzyme, thereby initiating transcription.
We have demonstrated σ54’s importance in regulating the expression of a wide range of natural products in Photorhabdus laumondii. This was achieved by utilizing the core domain of DctD (DctD(141-394)), a bEBP from Sinorhizobium meliloti. Further examination of the modified P. laumondi bEBP (mEBPs) revealed diverse effects on natural product expression involving activation and/or repression. Some of these mEBPs exhibit elevated expression levels and, in certain instances, demonstrate propensity for specific natural products compared to DctD(141-394). These preliminary findings suggest that mEBPs, could in time enable us to selectively express a subset of σ54-dependent BGCs. This would bypass the need for specific growth conditions, unlocking a reservoir of previously inaccessible natural products
Techniques to Enhance Just-In-Time Software Defect Prediction Models
No full textSoftware defects can lead to critical failures. Just-In-Time Software Defect Prediction (JIT-SDP) techniques identify potential defects early, improving software reliability and maintainability. This thesis addresses project clusters, data imbalance, and classifier combination challenges for JIT-SDP. The contributions were evaluated using diverse software projects and 34 datasets, totaling 259k commits.
The first contribution introduces ClusterCommit, a JIT-SDP approach tailored for project clusters sharing libraries and functionalities. Unlike traditional methods, ClusterCommit employs a machine learning model trained on commits from various projects within a cluster. The study incorporates six machine learning and three deep learning models. The results reveal noteworthy improvements, with mean Area Under the Curve (AUC) values ranging from 4% to 12%, particularly prominent in complex models such as Random Forest (RF) and Support Vector Machine (SVM) when dealing with large clusters. In contrast, simpler models like Naive Bayes (NB), Logistic Regression (LR), Decision Tree (DT), and k-Nearest Neighbors (k-NN) do not perform as well when applied to clusters of projects. This observed trend extends to deep learning models, where all models experience a performance from 3% to 30% with the ClusterCommit approach, irrespective of cluster size.
The second contribution proposed a One-Class Classification (OCC) approach to tackle the data imbalance challenge in JIT-SDP models. OCC algorithms, such as One-class SVM, Isolation Forest, and One-class k-NN, perform better than binary classifiers in medium to high data imbalance ratios. They achieve mean AUCs of 83%, 81%, and 86% for IOF, OC-k-NN, and OC-SVM, respectively, and require fewer features, reducing computational overhead.
Lastly, the JITBoost is a framework that uses a Boolean Combination of Classifiers to construct robust JIT-SDP models. Three BCC algorithms are investigated. JITBoost achieves superior performance by combining decisions from six traditional machine learning and deep learning algorithms, with mean AUCs of 89%, 87%, and 88% for JITBoost-BBC, JITBoost-IBC, and JITBoost-WPIBC, respectively
The Impact of Cranberry-Derived Polyphenols on Physical Performance and Skeletal Muscle Bioenergetics
Get PDFDietary choices have a direct impact on the gut microbiome, which in turn influences several body functions. In recent years, polyphenols, which are plant secondary compounds, have been shown to have prebiotic-like effects and associated with multiple health benefits. Cranberries are native to North America and have the highest polyphenol content and antioxidant capacity among the commonly consumed fruits and vegetables. Furthermore, cranberries stand out due to their high levels of the rare A-type proanthocyanidin (PAC-1), which is believed to be the main contributor to the beneficial effects. Recently, the concept that a link between skeletal muscle and the host’s gut microbiota exists was put forward, with several research groups proposing that supplementation with polyphenols could promote improved muscle function, and consequently, improve exercise performance. One proposed mechanism for the positive effects of polyphenols on muscle function is through improved mitochondrial capacity. Mitochondria are the main producers of ATP, and their ability to generate energy as efficiently as possible is directly related to performance, especially in endurance athletes. The effect of cranberry polyphenols on exercise performance and skeletal mitochondrial function has not been explored before. This dissertation consists of five chapters. Chapter 1 introduces key concepts to provide background information and states the rationale, objectives, and hypotheses of the dissertation. Chapter 2 describes a rodent study that aimed to investigate the effects of cranberry A-type proanthocyanidins combined with HIIT training on maximal running speed and skeletal muscle mitochondrial function. Chapter 3 consists of a systematic review with meta-analyses that synthesizes the current literature on the effects of polyphenol-rich berries on exercise performance, inflammation, and muscle damage. Chapter 4 is a clinical trial done with competitive/elite endurance runners that investigated the effects of a polyphenol-rich freeze-dried cranberry powder on running performance, lactate production, and skeletal muscle oxygenation. Chapter 5 is a follow-up study to the one described in Chapter 4 that aimed to explore the effects of the same cranberry powder on skeletal muscle mitochondrial capacity using near-infrared spectroscopy in healthy active adults. Finally, Chapter 6 discusses the findings from chapters 2-5 and provides general limitations and future research directions
Neuroimaging Fusion in Nonsubsampled Shearlet Domain by Maximizing the High-Frequency Subband Energy and Classification of Alzheimer's Disease using Local and Global Contextual CNN Features of Neuroimaging Data
Get PDFNeuroimaging fusion is the process of combining brain imaging data from multiple imaging modalities to create a composite image containing complementary information such as structural and functional changes in the brain. Recent advancements in transform domain fusion are promising, but challenges remain in accurately representing empirical distributions and maximizing energy in fused images. The most common neurodegenerative disease is Alzheimer’s disease, which demands accurate detection and classification for the care of the patient. Recent advancements in convolutional neural networks (CNNs)-based methods often overlook local features and do not pay attention to the discriminability of extracted features for the classification of Alzheimer’s disease. Moreover, existing architectures often end up using numerous parameters to enhance feature richness.
The objective of this thesis has two parts. In the first part, a novel statistically driven approach for fusing multimodal neuroimaging data is developed. In the second part, a lightweight deep CNN capable of extracting both local and global contextual features for the classification of Alzheimer’s disease is proposed.
In the first part of the thesis, a novel multimodal fusion algorithm using statistical properties of nonsubsampled shearlet transform coefficients and an energy maximization fusion rule is developed. The Student’s t probability density function is used to model heavy-tailed non-Gaussian statistics of empirical coefficients. This model is then employed to develop a maximum a posteriori estimator to obtain noise-free coefficients. Finally, a novel fusion rule is proposed for obtaining fused coefficients by maximizing the energy in the high-frequency subbands.
In the second part of the thesis, a novel lightweight deep CNN that extracts local and global contextual features for Alzheimer’s disease classification is proposed. The network is designed to process local and global features separately using specialized modules that enhance feature extraction relevant to the disease. Finally, the impact of fused images, obtained using the fusion approach of the first part, on the classification accuracy of Alzheimer’s disease is investigated.
Extensive experiments are carried out to validate the effectiveness of the various ideas and strategies proposed in this thesis for developing multimodal neuroimaging fusion and Alzheimer’s disease classification schemes