22691 research outputs found
Sort by
Comparing Optical and Interferometric Synthetic Aperture Radar-Based Remote Sensing Methodologies for Conflict Damage Assessment in Lebanon: A Case Study
Full text linkFAMILY HISTORY OF KIDNEY FAILURE, APOL-1 RISK VARIANTS, SOCIAL DETERMINANTS OF HEALTH, AND RISK OF CHRONIC KIDNEY DISEASE PROGRESSION: FINDINGS FROM THE CRIC STUDY
Full text linkKidney disease is often clustered within families, including Black families, and could be due in part to shared adverse social determinants of health (SDoH) and/or genetic factors. We examined the relationship between family history of kidney failure with SDoH and apolipoprotein L1 (APOL1) risk allele status, and the association between family history of kidney failure and chronic kidney disease (CKD) progression in people with CKD.
5,623 participants from Chronic Renal Insufficiency Cohort Study (CRIC), a longitudinal observational study, were used for this study. The exposure self-reported family history of kidney failure was defined as a first-degree relative treated for kidney failure with dialysis or transplantation. The outcome was CKD progression defined as incident end-stage kidney disease or 50% decline in estimated glomerular filtration rate (eGFR) from baseline. Logistic regression models were used to estimate adjusted odds ratios (aORs) of family history of kidney failure according to race-ethnicity/APOL1 risk allele status and SDoH. Cox proportional hazards models were used to assess the association of family history of kidney failure with the risk of CKD progression.
Among all participants (mean age 59.6 [SD 10.7] years; 43.7% female; 43.1% Black race), 948 (16.9%) reported a family history of kidney failure. Compared to White participants, Black participants were more likely to report a family history of kidney failure regardless of APOL1 status (aOR =2.25 (95% CI: 1.74-2.91) for 0 or 1 risk allele; and aOR=3.46 (95% CI: 2.39-5.02) for 2 risk alleles). Adverse SDoH, such as lower income and lower educational attainment, were positively associated with family history of kidney failure in crude models, but not in multivariable models. In prospective analysis, the family history of kidney failure was significantly associated with an increased risk of CKD progression in both crude (Hazard ratio [HR], 1.33 (95% CI: 1.19-1.49)) and multivariable models adjusting for demographics, APOL1 risk allele status, SDoH, and clinical factors (HR, 1.16 (95% CI: 1.02-1.33)).
These findings highlight the importance of collecting information on family history of kidney failure and further efforts to understand the reasons for familial aggregation of CKD
Playing Through the Future: On the Ethics of Advance Research Directives in Collision Sports Athletes
Full text linkChronic traumatic encephalopathy (CTE), a neurodegenerative disease caused by repetitive head injuries, poses significant concern for collision sports athletes. Since CTE and traumatic brain injuries (TBIs) can cause diminished decision-making capacity, researchers studying these conditions cannot always fulfill contemporary informed consent ethical and legal norms. Patient recruitment is challenging, stunting the development of supportive research for CTE and TBI treatment. Advance research directives (ARDs)—documents allowing pre-emptive indication of one’s research preferences in case of diminished decision-making capacity—can foster successful, ethical research. This paper primarily focuses on the meaningful use of ARDs in the U.S., offering a standard practice model to maximize implementation value by targeting collision sports athletes. Subsequently, this paper justifies this ARD model through public health ethics justificatory conditions. Future work could involve comparison of this model with other alternatives to the current surrogacy system
A Computational Framework To Reconstruct Dislocation Plasticity From Surface Measurement Data
No full textDislocations are fundamental to understanding the mechanical properties of crystalline materials, as they are the primary carriers of plastic deformation. The movement and interaction of dislocations under applied stress significantly influence material strength, ductility, and toughness. However, their microscale dimensions, complex three-dimensional distributions, and intermittent motion pose significant challenges for direct observation and characterization. Conventional techniques such as transmission electron microscopy, X-ray diffraction, and electron channeling contrast imaging provide valuable insights into dislocation structures. While these techniques can resolve the fine details of individual dislocations, they are limited in their ability to monitor dynamic dislocation evolution. Recent advancements in surface measurement techniques, including laser interferometry and acoustic emission sensing, offer high-frequency temporal resolution, making them promising candidates for studying dislocation dynamics. However, interpreting surface measurements into meaningful 3D dislocation microstructures remains a key challenge. This thesis develops a robust computational framework to reconstruct dislocation evolution from high-frequency surface measurement data, bridging the gap between surface responses and underlying dislocation activities.
At the core of this framework, an analytical solution for the three-dimensional elastodynamic field generated by dislocation motion is derived, demonstrating remarkable agreement between its predicted displacement and stress wave patterns and those obtained from molecular simulations. A scalable algorithm is designed to numerically implement this solution within dislocation dynamics simulations, enabling the computation of elastodynamic stress and displacement field at any point for evolving dislocations networks. To infer dislocation network evolution and reconstruct plastic strain localization, data assimilation techniques—including a total variation-regularized algorithm and the ensemble variational method—are employed. Rigorous assessments using simulation data demonstrate that the proposed approach achieves an accuracy exceeding 90% in three-dimensional reconstruction of plastic strain localization. The proposed computational framework advances the development of characterization techniques of crystalline materials and opens new avenues for understanding plastic deformation
ACTIVE SENSING AND TASK CONTROL IN TWO DISTINCT ANIMAL SENSORIMOTOR SYSTEMS
Full text linkMammals excel at performing complex motor actions, a capability that modern
robotic systems struggle to replicate. This ability arises from a combination of
feedback, feedforward, and active sensing control mechanisms. Feedback control uses external inputs and the current motor state to make adjustments toward a goal state. Feedforward control, on the other hand, maps a desired goal state to motor actions required to achieve the goal state. Active sensing allows animals to gather more information about their environment, often at the cost of goal-directed motor
actions, creating a trade-off between the two. This dissertation explores two distinct sensorimotor systems — visuomotor tracking in humans and echolocation regulation in bats — to understand how these three control mechanisms are implemented.
The first study examines people with cerebellar damage, which impairs motor
control. Using system identification techniques and a virtual reality tracking task, we investigate how feedforward and feedback pathways contribute to motor control and the cerebellum’s role in incorporating external stimuli into motion planning. Our findings show that cerebellar damage induces delays in both control pathways, but it does not alter the overall control dynamics. However, cerebellar damage significantly
impairs the ability to incorporate models of predictable stimuli to improve motion
control.
The second and third study investigate the active sensing behavior of bats
echolocating while hunting prey and characterize active sensing as a closed-loop control
process. They also examine the impact changes in goal states and environmental
factors have on echolocation behavior. In the second study, bats track a moving target in cluttered and uncluttered conditions. In the third study, we incorporate system identification techniques to quantitatively model the relationship between echolocation
changes and target motion by having the bats track oscillatory targets. Our results
demonstrate that bats dynamically adjust their echolocation calls in response to target movement, with environmental clutter inducing global shifts in call parameters. Bats also preferentially tune their echolocation to emphasize higher motion frequencies and to predict target motion.
The combination of these three experimental paradigms provides insights into the biological mechanisms underlying goal-directed motor control and the modulation and implementation of active sensing during motor control
INVESTIGATING Β-1,4-GALACTOSYLTRANSFERASE V KNOCKOUT IN HCT116 CELLS VIA CRISPR-CAS9: INSIGHTS INTO CELL PROLIFERATION AND CANCER PROGRESSION
No full textβ-1,4-Galactosyltransferase-V(GALT-V) is a glycosyltransferase enzyme involved in the synthesis of glycosphingolipids and glycoproteins, both of which are essential for regulating cell adhesion, signaling, and proliferation. Recent studies have identified GALT-V as significantly overexpressed at both the gene and protein levels as much as 6.5-fold increase in colorectal cancer (CRC) tissues compared to adjunct normal tissue suggesting it may play a role in tumor progression and metastasis. To evaluate its functional role, we performed CRISPR-Cas9-mediated knockout of GALT-V in HCT116 cells using high and low efficiency single guide RNAs (sgRNAs). GALT-V protein levels were reduced by 3.25 fold in the high-efficiency KO and 1.27 fold in the low-efficiency KO compared to untreated placebo cells as measured by ELISA. At the transcriptional level, qPCR revealed a 12-fold and 5.28-fold reduction in mRNA expression in high and low-efficiency KOs respectively. In addition, expression of key CRC-associated genes including Kirsten rat sarcoma viral oncogene homolog (KRAS), adenomatous polyposis coli (APC), and programmed death-ligand 1 (PD-L1) were analyzed following GALT-V knockout to assess downstream molecular effects. The high-efficiency guide resulted in a significant decrease in the mRNA levels of all three genes. Interestingly, the low-efficiency guide showed a greater reduction in APC mRNA levels compared to the high-efficiency knockout. This may be due to increased off-target effects from the low-efficiency guide, potentially leading to deletions in other regions of the genome. Additionally, cell viability and proliferation were measured using MTT assays, confluency tracking, immunofluorescence, and thymidine incorporation analysis—all of which showed significantly reduced GALT-V expression in the high-efficiency knockout cells. Both mRNA and protein levels of PD-L1 were reduced following GALT-V knockout, suggesting a potential role for GALT-V in regulating PD-L1 through n linked glycosylation affecting the stabilization of the protein. This study demonstrates that GALT-V promotes CRC cell survival and proliferation, potentially through modulation of KRAS and PD-L1. The downregulation of PD-L1 following GALT-V KO suggests there could be a mechanistic link between GALT-V mediated glycosylation and immune evasion in CRC. These findings highlight GALT-V as a promising molecular target for therapeutic intervention in colorectal cancer
Stem Cell-Driving Signals Activate Cell-Intrinsic Mesenchymal and Immunosuppressive Mechanisms via TGFBR2 in Glioblastoma
Full text linkHigh-grade gliomas, including glioblastoma (GBM), are highly heterogeneous with a complex oncogenic microenvironment consisting of distinct tumor niches and remarkable cellular variability. A critical component of GBM malignancy derives from the distinct population of multipotent and tumor-propagating glioma stem-like cells (GSCs), which maintain the vast and diverse cell landscape implicated in therapeutic inefficacy. Notably, attempts to activate an anti-tumor immune response in GBM have been met with many challenges due to its inherently immunosuppressive tumor microenvironment, particularly in mesenchymal-like tumors. The degree and mechanisms by which molecularly and phenotypically diverse GSCs contribute to this state are poorly defined. In this study, we describe a mechanism by which stem cell-driving events coordinate the transition to a mesenchymal-like GSC state through activation of TGFBR2 signaling. Furthermore, our multifaceted approach combining bioinformatics analyses of clinical and experimental datasets, single-cell sequencing, and molecular and pharmacologic manipulation of patient-derived cells identified GSCs expressing immunosuppressive effectors mimicking regulatory T cells (Tregs). We show that this Immunosuppressive Treg-Like (ITL) GSC state is specific to the mesenchymal GSC subset and is associated with and driven specifically by TGFb type II receptor (TGFBR2) in contrast to TGFBR1. Transgenic TGFBR2 expression in patient-derived GBM neurospheres promoted a mesenchymal transition and induced a 6-gene ITL signature consisting of CD274 (PD-L1), NT5E (CD73), ENTPD1 (CD39), LGALS1 (galectin-1), PDCD1LG2 (PD-L2), and TGFB1. This TGFBR2-driven ITL signature was identified in clinical GBM specimens, patient-derived GSCs and systemic mesenchymal malignancies. TGFBR2High GSCs inhibited CD4+ and CD8+ T cell viability and their capacity to kill GBM cells, effects reversed by pharmacologic and shRNA-based TGFBR2 inhibition. Collectively, our data identify a mesenchymal-like, immunosuppressive GSC state that is TGFBR2-dependent and susceptible to TGFBR2-targeted therapeutics. The impact of TGFBR2 inhibition on the anti-tumor immune response and the effects of combining TGFBR2-targeted treatment with current immunotherapy against GBM remain unknown. Nonetheless, our in vitro findings suggest that alternative treatment options to precisely target TGFBR2 should be pursued and prioritized moving forward
One VAE to Squeeze Them All: Causality-Aware Spatiotemporal Compression for Multi-Contrast MRI Latent Unification
No full textMedical image synthesis supports clinical decision-making by enabling tasks such as multi-modal contrast prediction, longitudinal tracking, and lesion characterization. In clinical practice, acquiring all MRI contrasts for each patient is often impractical due to time, cost, and patient burden, making the prediction of missing contrasts a valuable tool. Meanwhile, the heterogeneous nature of MRI data—with both static 3D and dynamic 4D images poses additional challenges for unified modeling. Recent advances in generative models, particularly Variational Autoencoders (VAEs), offer a promising direction for learning compact and informative latent representations from such complex data. In this work, we introduce a novel 3D/4D VAE-GAN framework for multi-contrast MRI image prediction and reconstruction. Our hybrid model preserves high-quality reconstruction and establishes a smooth latent space while dynamically adapting to varying temporal dimensions in MRI sequences. Both quantitative and qualitative evaluations confirm excellent reconstruction quality and clear separation of different lesion subtypes, demonstrating the model’s ability to capture clinically relevant spatiotemporal features. Furthermore, the compressed latent representations can be integrated into diffusion models for image synthesis and missing contrast prediction, and facilitate downstream tasks like lesion segmentation, patient retrieval, and enabling privacy-preserving data sharing. These findings also offers a robust foundation for future data-driven healthcare applications
Enabling Conditional Generation Without Training: Plug & Play Generative Modelling using Diffusion Models
No full textDiffusion models have achieved state-of-the-art results across various generative modeling tasks, including text, image, video, and audio generation. By iteratively refining standard Gaussian noise into structured signals, these models offer remarkable flexibility. However, their practical capabilities remain constrained by the limited availability of large-scale annotated datasets, particularly for complex and compositional tasks that require adherence to specific rules. These limitations restrict their applicability to relatively simple problems with smaller datasets. In this thesis, we propose a series of methods to reduce the dependency on large-scale data, enabling diffusion models to solve complex inverse problems more effectively. First, we introduce Unite and Conquer, a fully diffusion-based framework that decomposes complex compositional tasks into simpler subproblems, reducing data requirements by a logarithmic factor at the cost of increased computation. By leveraging the product of experts approach from probability theory, we train individual diffusion models to solve these simpler subproblems and then combine their capabilities during inference, enabling the resolution of complex tasks without requiring extensive annotated datasets. Second, we eliminate the need for training multiple individual diffusion models by utilizing a single unconditional diffusion model. By integrating problem-specific physical constraints and inverse dynamics, we introduce Steered Diffusion, which replaces the need for separate models with optimization iterations embedded within the generation process. Third, we enhance Steered Diffusion with DreamGuider, a method that removes its memory overhead and reliance on handcrafted parameters. By incorporating zeroth-order optimization and automatic parameter estimation, DreamGuider enables conditional generation with minimal computational overhead, eliminating manual tuning while improving performance. Finally, we introduce MaxFusion, a method for compositional and controllable text-to-image generation in diffusion models. MaxFusion enhances controllability in complex scenarios where multiple 3D and 2D conditioning inputs must be simultaneously incorporated. Collectively, these contributions advance the conditional generation capabilities of diffusion models, reducing their dependence on large-scale annotated datasets and enabling physics-based, interpretable, and controllable generation
te ‘ori tahiti e te ‘ahu: Public Performances of Tahitian Dance and Costuming Innovations During the Height of French Colonialism in Tahiti, 1880-1910
Full text linkIn Tahiti in 1956, Madeleine Moua founded the Heiva dance troupe and made bold costuming changes, replacing the modest Western clothes worn throughout the last century of dance. Historians and cultural scholars attribute this moment as one of the central components that kickstarted the indigenous mā’ohi cultural revival of the second half of the 20th century. What do scholars miss in skipping over the cultural innovations exhibited during the colonial period? While Moua’s contributions to Tahitian dance attire and public performance are undoubtedly important, the notion that Moua was the first to lay claim to “authentic” indigenous dance practices in the “post-colonial” period obscures the creative cultural activism of the early colonial period from 1880-1910. This thesis makes this visible by examining Tahitian dance costuming, the externalization of Tahitian dance as a symbol of cultural and political identity, and the public performance of various types of ‘ori tahiti (Tahitian dance) at a time when it was not perceived as decent by authorities, many indigenous elites, and even lower-class locals. Building on debates about the invention of tradition (Hobsbawm, Linnekin, Plant, etc.) and anthropological work on Tahiti and Hawai’i (Tcherkézoff, Imada, etc.), this research problematizes the scholarly assumption that Tahitian material innovation and public dancing did not occur under French colonial rule at the turn of the 20th century. Instead, this thesis demonstrates how the Fête de Tiurai helped popularize Tahitian dance and costuming innovations decades before the dance and cultural revivals of the 1950s-1970s. Additionally, while Hawai’ian dance performances abroad were more frequent in the period from 1880-1910, a one-of-a-kind Tahitian dance troupe from 1906 centers Tahiti in a field where the mā’ohi are often lumped in with the sexualization of Hawai’ian hula or the pan-Polynesian experience. Despite Western feminization of Polynesian cultures and the theatricalization of cultural performances, Tahitians at the turn of the 20th century made their mark on their indigenous cultural traditions, their innovations paving the way for Moua and cultural “revivals” decades later