22689 research outputs found
Sort by
MANAGING ERRORS IN QUANTUM METROLOGY VIA QUANTUM ERROR MITIGATION
Full text linkQuantum sensing is an emerging field of quantum technology, which has already
seen practical application and shows promise for future development. However, quantum sensors, like all quantum devices, are plagued by noise, which limits their performance. This dissertation develops a theoretical framework for mitigating noise
in quantum sensing protocols, by leveraging techniques from quantum error mitigation and quantum control. In chapters 1 and 2, we begin by reviewing the basics of quantum sensing and noise, including the theory of open quantum systems, zero-noise extrapolation, and the filter function formalism. We then apply these basic frameworks to the problem of interest. In chapter 3, we consider the case of mitigating Markovian noise, studying the application of quantum error mitigation to Ramsey interferometry. Theoretical results are derived demonstrating the effectiveness of zero-noise extrapolation in reducing the effect of Markovian noise to improve device performance. In chapter 4, we move on to discuss the case of non-Markovian correlated noise. By combining zero-noise extrapolation
with elements from quantum control and the filter function formalism, we once again demonstrate the effectiveness of zero-noise extrapolation in mitigating device noise—this time, correlated noise, beyond the scope for which zero-noise extrapolation was originally developed.
In addition to demonstrating performance improvements, this cross-pollination
of techniques from error mitigation and quantum control suggests a broad avenue for future research and development of hybrid strategies. By unifying concepts from quantum error mitigation and advanced control protocols, we open new pathways
toward more robust and adaptable quantum sensors capable of operating in a variety of noise environments. This union not only enhances the capabilities of current-generation devices but also provides a theoretical foundation for addressing noise processes in
near-term and future devices. As such, the work presented here lays the groundwork for comprehensive, multi-faceted error management approaches that could ultimately accelerate the maturation of quantum sensing technologies
Sumerian Proverb Collection 2
No full textIt is often difficult or impossible to translate Sumerian proverbs accurately. Nevertheless, they are commonly cited in Assyriological literature as evidence for various arguments. Improvements on their first editions are selective and infrequently published, but thanks to progress in the fields of Sumerian language and literary study, many can now be revised and convincingly understood. Where they still evade explanation, it is often the case
that new evidence greatly reduces previous speculation.
Sumerian proverb collections were not intended to be enjoyed by a modern audience. In fact, as far as the archaeological evidence can prove, they were not intended to be enjoyed very far outside of the classroom at all. In translation, they seem like bare lists of hundreds of phrases, but in Sumerian their sequencing is restlessly clever, and it becomes clear that such collections were designed not only to teach cuneiform but to teach appreciation for the powers of literary expression. In terms of genre, the collections were neither lists nor literature, but to a special and limited extent, both; they were lists-at-play.
The following is an updated edition of Sumerian Proverb Collection 2. Gordon’s
pioneering editio princeps of SP 1 and 2 was made available in 1959. Alster finally published the entire proverb corpus in 1997 but could not devote much space to any specific collection. In many cases, translations were based on guesswork, and can be improved with new databases, glossaries, and grammars. In addition, dozens of fragment identifications, joins, and analyses have filled in crucial gaps since 1997. The resulting translations usually have entirely new meanings.
This study is presented in three sections. The first is an orienting discussion of SP 2 in three chapters: an introduction, a description of the curriculum, and a description of the text. The second section is the new edition, including a lengthy commentary addressing each entry in the collection. Finally, source descriptions and a bibliography are provided for reference. It is hoped that this updated translation better conveys the ancient sayings and provides more
secure evidence from which other researchers may draw new conclusions
EXPLORING PREFERENCES FOR TUBERCULOSIS CONTACT INVESTIGATION STRATEGIES AMONG INDIVIDUALS WITH TB AND HOUSEHOLD CONTACT PERSONS IN SOUTH AFRICA: A DISCRETE CHOICE EXPERIMENT
Full text linkTuberculosis (TB) remains one of the leading causes of death worldwide. However, household contact investigation has not been widely implemented in low- and middle-income countries, which bear a high burden of TB. Understanding the preferences and needs of those affected by TB is crucial for designing effective contact investigation strategies, optimizing resource distribution, and increasing the uptake of contact investigations.
We conducted a cross-sectional survey-based discrete choice experiment in South Africa to quantify the preferences of individuals with TB and their household contacts regarding six key domains of household contact investigation. Participants were presented with eight question sets, each offering two mutually exclusive options. Our results indicated that the provision of incentives and the option to undergo hypertension and diabetes screening during contact investigation visits are strong drivers for choice. Several key attributes also varied between rural and urban areas. Rural residents favored visits during off-hours, including early evening, weekends, and holidays, while urban residents preferred visits during routine working hours. In rural areas, plain clothes and unmarked cars that conceal the identities of healthcare workers were preferred over uniforms and marked cars, whereas urban residents had the opposite preference. Participants showed little interest in undergoing HIV testing during the visit, likely due to the stigma associated with HIV.
Our study addressed the research gap by examining the preferences for in-person TB contact investigation strategies among individuals directly affected by TB in South Africa. Understanding these preferences enables us to design more acceptable and successful contact investigation interventions tailored to target populations, setting valuable guidance for future contact investigation implementation
The Optimization of Multicolor Immunofluorescence Panels to Deeply Immunophenotype Monocytes from People with Multiple Sclerosis
Full text linkMultiple sclerosis (MS) is a chronic autoimmune disease characterized by inflammation, demyelination, and neurodegeneration in the central nervous system (CNS). Its pathogenesis involves a complex interplay between immune cells, with monocytes increasingly recognized as key mediators of both inflammatory damage and repair processes. Precision medicine approaches in MS aim to define patient-specific immunological profiles to guide tailored interventions, highlighting the importance of high-resolution immune monitoring tools.
Monocytes are functionally heterogeneous and classified into classical (CD14++CD16−), intermediate (CD14++CD16+), and non-classical (CD14+CD16++) subsets, including 6-sulfo LacNAc+ (slan) and slan− non-classical populations. These subsets differ in their migratory capacities, effector functions, and receptor expression profiles. In MS, monocyte-driven mechanisms such as heterotypic aggregation, antigen presentation, and dysregulated expression of chemokine, adhesion, scavenger, Fc, and checkpoint receptors contribute to disease progression. Chemokine receptors including CCR2, CXCR4, and CX3CR1 direct monocyte trafficking into the CNS, where they amplify inflammation or, conversely, support tissue remodeling depending on environmental cues.
To enable comprehensive profiling of monocyte diversity and function, optimized multicolor immunofluorescence panels (OMIPs) are essential. This thesis incorporates OMIP-083, a 21-marker 18-color flow cytometry panel specifically designed to characterize human monocyte subsets, developed by Hally et al. We report further optimization of OMIP-083 to improve resolution of rare subsets such as SLAN+ non-classical monocytes and to assess functional markers including PD-1, CD86, CD36, and CD11b. The use of this optimized panel in cryopreserved peripheral blood mononuclear cells (PBMCs) is one of if not the primary adaptation of this work.
By integrating detailed immunophenotyping with a focus on differential monocyte function, this thesis supports the development of categorized biomarker strategies for MS and offers a model for translating high-dimensional immune data into clinically relevant insights. Ultimately, targeting monocyte-driven pathways personalized to the profile of the patient may enhance therapeutic precision in MS, particularly in patients with progressive or treatment-resistant forms of the disease
A CASE FOR REPRESENTATION IN THE PERCEPTION OF ROTATION
Full text linkObjects have privileged role in visual processing. One of the reasons is that we often perceive individuals that do not qualify as conventional objects through principles of object-centered representation. In this dissertation, I propose that orbits are represented like objects and develop the orbit-centered representation theory. The orbit-centered representation predicts that tracking a target, for example a disk in a two-disk orbiting system, should be deficient compared to tracking a target in highly similar but non-rotational motion. Using a multiple object tracking task, this prediction is confirmed in the first empirical study. Then I apply this framework to explain the motion silencing illusion in the second empirical study. I reason that motion silencing is an attribution error caused by implicit tracking failure rather than a general processing limitation. I demonstrate that rotational motion produces stronger motion silencing illusions than controlled non-rotational motion across various conditions. The results confirm the prediction from the orbit-centered representation, specifically tracking in rotation is deficient. In the third empirical study, I show the benefit of the orbit-centered representation. If the orbit format of representation is symmetric to the computations needed to carry out the task, performance is enhanced compared to a representational format that conflicts with the task. Specifically, a group of disks arranged on the perimeter of a square, tracking one disk when the disks are rotating together around the square center shows higher accuracy than tracking one disk when they are translating on the perimeter of the square. I argue that the orbit format elicited in rotation is a convenient representation for this task while individuals translating violates the object centered representation caused by disks aligned nicely on the square. Taken together, by examining the orbit-centered representation in visual perception, this dissertation provides evidence supporting a view of perception that is strongly representational. A strong representation includes properties that are not present in the low-level stimulus like axes and relational variables. This dissertation also underscores on explaining the errors and illusions in visual perception as how the visual input is represented rather than bottleneck or resource limitations in the visual system
THRESHOLD-FREE NEURAL NETWORK MODELS FOR DETECTION OF ZEBRAFISH LARVAE BOUTS FROM TAIL ANGLE KINEMATICS
No full textZebrafish larvae are a powerful model for studying sensorimotor processing, often utilizing the optomotor response (OMR). Accurately detecting discrete swimming bouts- the fundamental units of zebrafish locomotion- from kinematic data is crucial for behavioral analysis. However, traditional threshold-based detection methods struggle with noise and variability in bout kinematics. This thesis addresses the need for robust, threshold-free bout detection by developing and evaluating deep learning models based on tail angle kinematics. We present two convolutional neural network (CNN)-based models: an offline CNN-LSTM model for post-processing analysis and a streamlined online CNN model for real-time applications. These models were trained and validated using tail tracking data acquired from head-fixed larval zebrafish subjected to OMR stimuli via a novel, integrated Stytra+PsychoPy closed-loop workflow, which overcomes stimulus refresh rate limitations of default setups. Both models achieved high performance without relying on pre-defined thresholds; the offline model yielded mean precision/recall of 83.15%/89.83%, while the online model demonstrated viability for real-time use with minimal performance decrease (mean precision/recall 84.09%/87.14%), albeit with measurable latency. We validated our experimental setup by showing that bout kinematics in head-fixed fish under closed-loop conditions were largely comparable to free-swimming behavior, supporting the biological relevance of the training data. Furthermore, we applied the offline model to compare OMR kinematics across wild-type (WT) AB, Nacre, and Casper strains. While Nacre and Casper exhibited similar kinematics, both differed significantly from WT AB, particularly in bout duration under closed-loop conditions and in adaptation dynamics during closed-loop/open-loop transitions. These findings highlight the effectiveness of threshold-free deep learning models for zebrafish bout detection and underscore the importance of considering strain-specific behavioral differences in sensorimotor research
CLIMATOLOGY OF ATLANTIC TROPICAL CYCLONES AND MEDIA RESPONSE TO HURRICANE MILTON: LINKS TO CLIMATE CHANGE
Full text linkMedia response to Hurricane Milton in the fall of 2024 led to varying and even contrasting opinions on how anthropogenic climate change influences tropical cyclone activity. This research delves into the political biases between groups of media outlets and how they respond to Hurricane Milton when it comes to tropical weather’s relationship to climate change. Background literature on climate change and tropical weather suggests a relationship which must be communicated effectively to the public to limit political bias or reader confusion. Analysis includes observational and forecast data throughout Hurricane Milton’s life, plotted alongside media article responses to find relationships between the two and with the media’s political bias. The categorization of certain news outlets and their inherent political bias is supported by AllSides Media. It is important to understand the media’s response to tropical weather because of the profound influence and reach of this platform over the public. A sound and well-established relationship between the atmospheric scientific community and the mass media is essential for adequate communication to the public regarding extreme meteorological events and the influence of climate change
TARGETING HPV-INDUCED VULNERABILITIES FOR TUMOR-SPECIFIC SENSITIZATION
No full textCervical cancer is ranked fourth among women’s cancers and is a public health issue (Caruso
et al., 2024). Each year, more than 660,000 women are diagnosed with cervical cancer,
and over 348,000 lose their lives to the disease (Sahasrabuddhe, 2024). One of the main
causative agents for cervical cancer are high-risk Human Papilloma Viruses (HPV) that
induce oncogenesis through two of its viral proteins E6 and E7. E6 induces various cell
changes, from reprogramming the epigenome to inhibiting tumor suppressors like p53. Locally
advanced cervical cancer is treated with concurrent radiation and chemotherapy. However,
not all tumors respond well to this modality, indicating a need to improve the efficacy of
radiotherapy in HPV-positive cervical cancer.
We have uncovered that disrupting the centromeric protein complex MIS12 by targeting its
member NSL1 specifically sensitizes E6-expressing cells to radiation. Based on this finding, we
hypothesized that Aurora Kinase (AurkB) inhibitor (AZD1152) facilitates radio-sensitization,
specifically in HPV-positive cells, since AurkB is required for the mitotic function of the
MIS12 complex during mitosis. SiHa cells are HPV-positive patient derived cell line which
has two copies of E6 and E7 oncoprotein. We have found that AurkB inhibition in SiHa WT
cells sensitize them to Ionizing radiation (IR). We performed knock down of other members
of MIS12 complex in SiHa WT cells to determine if this induces a similar effect like NSL KO.
We see a trend that Knockdown of MIS12 sensitizes SiHa WT cells to IR at certain doses but
not in SiHa E6 KO cells, this might suggest that the sensitization to MIS12 depletion is E6
dependent. However, this still needs to be validated. In a reciprocal approach, we overexpress
E6 in HPV-negative cell lines to study its effect on radio-sensitization. We have generated two new cell lines, breast cancer cell line MCF7 and HPV-negative cervical cancer cell line
C33A where we have stably overexpressed eGFP or E6 through lentiviral transduction. We
further plan to test the inhibitor in these cell lines too.
Ultimately, we aim to understand the underlying mechanism of radio-sensitization when
Centromeric Protein Complex is disrupted and why this sensitization is E6 dependent.
Through these findings, we hope to develop a combined therapeutic approach consisting of
concurrent radiotherapy and AurkB inhibitor to improve treatment outcomes in cervical
cancer patients
ARACHNOBOT: DEVELOPING A ROBOT CAPABLE OF TRAVERSING NETTING
No full textRobots struggle to traverse netting upside down (e.g., fishing nets, climbing nets). Current literature does not have much information on robots walking on netting, such as an artificial web. On the other hand, spiders are incredible at maneuvering webs. We set an objective to create a spider-inspired robot able to traverse an artificial web (a form of netting) upside down, and explore what factors are important in traversing an artificial web. Our biological inspiration and reference was the spider Uloborus diversus. We tried a novel robot gait, which was using the limb, bearing the most body weight, to pull the body forward. Trials with our robot gait led us to realize an important guideline when traversing an artificial web upside down: only one limb should have its torque engaged when attempting to move the robot’s center of mass (CoM). An individual limb is not strong enough to move the robot’s center of mass. Coordinated movement amongst all the limbs is required, which involves releasing all limb torques except for the one limb pulling the robot’s body across the artificial web. In other words, our robot needs to swing its center of mass to traverse the web using the force of a single limb. In addition to the exploration of spider-inspired robophysics, we also developed a 3D visual analysis pipeline to process spider locomotion videos, generating tracked 3D locomotion
Shell Microelectrode Arrays for 3D Electrophysiological Characterization of Cardiac Organoids
No full textHuman cardiac organoids have emerged as promising models for investigating cardiogenesis and cardiovascular diseases. These 3D structures reproduce key features of cardiac tissue organization and function, especially the electrophysiology patterns of the heart, which allows the examination of cardiac electrical activities using artificial heart-like tissue while providing advantages over conventional 2D cardiac tissue cultures. Yet, conventional electrophysiological assessment methods are limited to 2D measurements that fail to capture the spatiotemporal complexity of 3D electrical propagation. This thesis presents the 3D mapping of electrophysiology propagation of cardiac organoids with a novel conformal shell microelectrode array (MEA) system with 16 evenly positioned electrodes. Using the shell MEA, we achieved a higher signal-to-noise ratio compared to 2D recording systems and consistent detection of activation time. Radial basis function fitting was applied to perform 3D spatiotemporal mapping and visualization on the surface of the cardiac organoid, and the finite differential equation was used to calculate the conduction velocity. The accuracy of the mapping was validated with calcium imaging with a high correlation (>0.94) between the 3D isochrone maps generated by the shell MEAs and the calcium transient propagation patterns observed through fluorescence microscopy. Furthermore, we demonstrate the capability of long-term recordings and pharmacological testing using cardioactive compounds such as isoproterenol, serotonin, and E-4031, with observed changes in spike-shape parameters aligning with expected drug-induced effects