UTSA Runner Research Press (Univ. of Texas at San Antonio)
Not a member yet
6846 research outputs found
Sort by
QWR-Dec-Net: A Quaternion-Wavelet Retinex Framework for Low-Light Image Enhancement with Applications to Remote Sensing
Full text linkComputer vision and deep learning are essential in diverse fields such as autonomous driving, medical imaging, face recognition, and object detection. However, enhancing low-light remote sensing images remains challenging for both research and real-world applications. Low illumination degrades image quality due to sensor limitations and environmental factors, weakening visual fidelity and reducing performance in vision tasks. Common issues such as insufficient lighting, backlighting, and limited exposure create low contrast, heavy shadows, and poor visibility, particularly at night. We propose QWR-Dec-Net, a quaternion-based Retinex decomposition network tailored for low-light image enhancement. QWR-Dec-Net consists of two key modules: a decomposition module that separates illumination and reflectance, and a denoising module that fuses a quaternion holistic color representation with wavelet multi-frequency information. This structure jointly improves color constancy and noise suppression. Experiments on low-light remote sensing datasets (LSCIDMR and UCMerced) show that QWR-Dec-Net outperforms current methods in PSNR, SSIM, LPIPS, and classification accuracy. The model’s accurate illumination estimation and stable reflectance make it well-suited for remote sensing tasks such as object detection, video surveillance, precision agriculture, and autonomous navigation
Solar Photovoltaic System Fault Classification via Hierarchical Deep Learning with Imbalanced Multi-Class Thermal Dataset
Full text linkThe growing global reliance on solar photovoltaic (PV) systems requires robust, automated inspection techniques to ensure reliability and efficiency. Thermal infrared (IR) imaging is widely used for detecting PV faults; however, accurate classification remains challenging due to severe class imbalance, low thermal contrast, and high inter-class visual similarity among fault types. This study proposes a hierarchical deep learning framework for thermal PV fault classification, integrating a multi-class dataset-balancing strategy to enhance representational efficiency. The proposed framework consists of two major components: (i) a hierarchical two-stage classification scheme that mitigates data imbalance and leverages limited labeled data for improved fault discrimination; and (ii) a contrast-preserving MixUp augmentation technique designed explicitly for low-contrast thermal imagery, improving minority fault class recognition and overall robustness. Comprehensive experiments were conducted on benchmark 8-class thermal PV datasets using nine deep network architectures. Dataset refactoring decisions are validated through quantitative inter-class distance analysis using multiple complementary metrics. Results demonstrate that the proposed hierarchical SlantNet model achieves the best trade-off between accuracy and computational efficiency, achieving an F1-Efficiency Index of 337.6 and processing 42,072 images per second on a GPU, over twice the efficiency of conventional approaches. Comparatively, the Swin-T Transformer attained the highest classification accuracy of 89.48% and F1 score of 80.50%, while SlantNet achieved 86.15% accuracy and 73.03% F1 score with substantially higher inference speed, highlighting its real-time potential. Ablation studies on augmentation and regularization strategies confirm that the proposed techniques significantly improve minority class detection without compromising overall performance, with detailed per-class precision, recall, and F1 analysis. The proposed framework delivers a high-accuracy, low-latency, and edge-deployable solution for automated PV inspection, facilitating seamless integration into operational PV plants for real-time fault diagnosis
From Molecules to Medicine: Deciphering Obesity and Lipid Metabolism for Translational Insights
Full text linkObesity, type 2 diabetes (T2D), and insulin resistance are pervasive metabolic disorders marked by chronic low-grade inflammation and systemic metabolic disorders. The emerging field of immunometabolism highlights how interactions between immune processes and metabolic pathways in adipose tissue, liver, muscle, and pancreatic islets contribute to disease pathogenesis. Lipid dysregulation plays a central role in these processes, with distinct lipid molecules identified in obese patients as compared to lean patients that correlate with insulin resistance, inflammation, and vascular dysfunction. This Special Issue compiles a multidisciplinary body of research aimed at elucidating molecular mechanisms, identifying novel biomarkers, and exploring innovative therapeutic strategies. Key contributions include studies on omega-3 long-chain polyunsaturated fatty acids (LCPUFAs) and their differential associations with neurocognitive development; the potential of beta-defensin 2 as a biomarker linking gut-derived inflammation and metabolic dysfunction; and the promotion of adipocyte browning by Carnosic acid via AMPK activation and GSK3β inhibition. Additionally, reviews of phytochemicals underscore their multisystem therapeutic potential, while investigations into sodium–glucose cotransporter-2 (SGLT2) inhibitors suggest possible metabolic and neuroprotective benefits beyond glucose control. Maternal lipid metabolism during pregnancy and its impact on maternal fetal health further emphasize the clinical complexity of lipid dysregulation. Despite promising insights, significant gaps remain regarding causality versus correlation in lipid biomarkers, standardization of analytical methodologies, tissue heterogeneity, and unintended effects of metabolic interventions. Collectively, these studies underscore the necessity of integrative, mechanism-driven research to bridge fundamental biology with translational and clinical applications, ultimately advancing precision therapies for metabolic diseases
H-Wave® Device Stimulation for Chronic Knee Pain Disorders: A Patient-Reported Outcome Measures Observational Study
Full text link<i>Background and Objectives</i>: Chronic knee pain (cKP) affects approximately 25% of adults worldwide, with prevalence increasing over recent decades. While conventional treatments have clinical limitations, several types of electrical stimulation have been suggested to improve patients&rsquo; quality of life. The electrical stimulation literature contains inadequate patient-reported outcome measures (PROMs) data. Encouraging preliminary H-Wave<sup>&reg;</sup> device PROMs results for chronic neck, shoulder, and low back pain have previously been published. This PROMs study&rsquo;s goal is to similarly assess the efficacy of H-Wave<sup>&reg;</sup> device stimulation (HWDS) in patients with differing knee disorders. <i>Materials and Methods</i>: This is an independent, retrospective, observational cohort study analyzing H-Wave<sup>&reg;</sup> PROMs data, prospectively and sequentially collected over 4 years. In total, 34,192 pain management patient final surveys were screened for participants who were at least 18 years old, used H-Wave<sup>&reg;</sup> for any knee-related disorder, reporting chronic pain from 90 to 730 days, with device treatment duration from 22 to 365 days. PROMs included effects on function, pain, sleep quality, need for medications, ability to work, and patient satisfaction; additional data includes gender, age (when injured), chronicity of pain, prior treatments, and frequency and length of device use. <i>Results</i>: PROMs surveys from 34,192 HWDS patients included 1143 with &ldquo;all knee&rdquo;, 985 &ldquo;knee injury&rdquo;, and 124 &ldquo;knee degeneration&rdquo; diagnoses. Reported improvements in function/ADL (96.51%) and work performance (84.63%) were significant (<i>p</i> &lt; 0.0001), with &ge;20% pain relief in 86.76% (<i>p</i> &lt; 0.0001), improving 2.96 points (average 0&ndash;10 NRS). Medication use decreased (69.85%, <i>p</i> = 0.0008), while sleep improved (55.33%) in knee injury patients. Patient satisfaction measures exceeded 96% (<i>p</i> &lt; 0.0001). Subgroup analysis suggests that longer device use and shorter pain chronicity resulted in increased (<i>p</i> &lt; 0.0001) HWDS benefits. <i>Conclusions</i>: HWDS PROMs data analysis demonstrated similarly encouraging outcomes for cKP patients, as previously reported for several other body regions. Knee injury and degeneration subgroups had near-equivalent benefits, as observed for all knee conditions. Despite many reported methodological limitations, which limit causal inference and preclude broader recommendations, HWDS appears to potentially offer several benefits for refractory cKP patients, requiring further studies
Race, Breastfeeding Support, and the U.S. Infant Formula Shortage: An Exploratory Cross-Sectional Study
Full text link<b>Background/Objectives:</b> African American women are less likely to breastfeed in general and to breastfeed exclusively for the first six months of infancy. Racial and ethnic breastfeeding disparities are especially pronounced in the South, particularly in rural communities. These differences are attributed largely to structural lactation impediments that include less breastfeeding support in healthcare settings, workplaces, and communities. While a great deal of research has explored racial differences in breastfeeding, minimal attention has been paid to the social correlates and racial disparities associated with the 2022 U.S. infant formula shortage. Our study explores racial distinctions in the formula shortage&rsquo;s effect on breastfeeding support among Gulf Coast Mississippians. <b>Methods:</b> We use data from the second wave of the Mississippi REACH (Racial and Ethnic Approaches to Community Health) Social Climate Survey to determine if racial differences are evident in the formula shortage&rsquo;s influence on breastfeeding support. We predict that the infant formula shortage will have prompted African American respondents to become much more supportive of breastfeeding than their White counterparts, net of sociodemographic controls. This hypothesis is based on the lower prevalence of exclusive breastfeeding among African Americans, thereby indicating a greater reliance on formula. The study uses a general population (random digit dial) sample and purposive (exclusively African American) oversample to analyze validated data from a cross-sectional survey. Sampling took place between September and December 2023, with a sample population of adult male and female Mississippians. A series of binary logistic regression models were employed to measure the association of race with breastfeeding support changes resulting from the infant formula shortage. <b>Results:</b> The study results support the hypothesis, as seen by a positive association between African Americans and increased breastfeeding support directly related to the infant formula shortage. Further, the baseline statistical model reveals African American respondents to be five times more likely than White respondents (<i>p</i> &lt; 0.001) to report that the formula shortage increased their support of breastfeeding. <b>Conclusions:</b> We conclude by discussing this study&rsquo;s implications and promising directions for future research
Fast Computation for Square Matrix Factorization
Full text linkIn this work, we discuss a method for the QR-factorization of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>N</mi><mo>&times;</mo><mi>N</mi></mrow></semantics></math></inline-formula> matrices where <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>N</mi><mo>&ge;</mo><mn>3</mn></mrow></semantics></math></inline-formula> which is based on transformations which are called discrete signal-induced heap transformations (DsiHTs). These transformations are generated by given signals and can be composed by elementary rotations. The data processing order, or the path of the transformations, is an important characteristic of it, and the correct choice of such paths can lead to a significant reduction in the operation when calculating the factorization for large matrices. Such paths are called fast paths of the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>N</mi></mrow></semantics></math></inline-formula>-point DsiHTs, and they define sparse matrices with more zero coefficients than when calculating QR-factorization in the traditional path, that is, when processing data in the natural order <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><msub><mrow><mi>x</mi></mrow><mrow><mn>0</mn></mrow></msub><mo>,</mo><msub><mrow><mi>x</mi></mrow><mrow><mn>1</mn></mrow></msub><mo>,</mo><msub><mrow><mi>x</mi></mrow><mrow><mn>2</mn></mrow></msub><mo>,</mo><mo>&hellip;</mo></mrow></semantics></math></inline-formula>. For example, in the first stage of the factorization of a 512 <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>&times;</mo></mrow></semantics></math></inline-formula> 512 matrix, a matrix is used with 257,024 zero coefficients out of a total of 262,144 coefficients when using the fast paths. For comparison, the calculations in the natural order require a 512 <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>&times;</mo></mrow></semantics></math></inline-formula> 512 matrix with only 130,305 zero coefficients at this stage. The Householder reflection matrix has no zero coefficients. The number of multiplication operations for the QR-factorization by the fast DsiHTs is more than 40 times smaller than when using the Householder reflections and 20 times smaller when using DsiHTs with the natural paths. Examples with the 4 &times; 4, 5 &times; 5, and 8 &times; 8 matrices are described in detail. The concept of complex DsiHT with fast paths is also described and applied in the QR-factorization of complex square matrices. An example of the QR-factorization of a 256 <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>&times;</mo></mrow></semantics></math></inline-formula> 256 complex matrix is also described and compared with the method of Householder reflections which is used in programming language MATLAB R2024b
A Two-Stage Feature Screening Framework for Ultrahigh-Dimensional Survival Data
Full text linkIdentifying important features associated with right-censored survival time in ultrahigh-dimensional survival data is a challenging task due to the curse of dimensionality and information loss caused by censoring. To address these challenges, we propose a two-stage feature screening framework consisting of an imputation step and a feature screening step. The use of Buckley&ndash;James imputation leverages information from censoring and can therefore enhance the overall screening performance, particularly when the censoring rate is relatively high. We establish the sure screening properties of the two screening procedures proposed under this framework and illustrate their advantages through simulations. A real-world example is also provided to demonstrate the practical usefulness of the proposed approach
Computing the Dissociation Constant from Molecular Dynamics Simulations with Corrections for the Large Pressure Fluctuations&mdash;Aquaglyceroporins Have High Affinity for Their Substrate Glycerol
Full text linkIn this paper, we consider the inevitable large fluctuations of pressure in typical molecular dynamics (MD) simulations of ligand&ndash;protein binding problems. In simulations under the constant pressure of one bar, the pressure artifactually fluctuates over the range of <inline-formula><math display="inline"><semantics><mrow><mo>&plusmn;</mo><mn>100</mn></mrow></semantics></math></inline-formula> bars or more. This artifact can cause gross inaccuracy in the apparent binding affinity computed as the ratio of the probability for the ligand to be bound inside the protein and the probability for the ligand to be outside the protein. Based on statistical thermodynamics, we derive a correction factor for the ligand&ndash;protein binding affinity to compensate for the artifactual pressure fluctuations. The correction factor depends on the change in the system volume between the bound and the unbound states of the ligand. We conducted four sets of MD simulations for glycerol affinities with four aquaglyceroporins: AQP10, AQP3, AQP7, and GlpF. Without the correction factor, the apparent affinity of glycerol with each of these four aquaglyceroporins is computed directly from the simulations to be very low (~1/M). With the correction factor applied, glycerol&rsquo;s affinity is computed to be 1/mM to 1/&micro;M. In conclusion, glycerol has high affinity for its native facilitator aquaglyceroporins, which is in contrast to the current literature not correcting the artifactual consequences of the large pressure fluctuations in typical in silico experiments
Electronic and Optical Behaviors of Platinum (Pt) Nanoparticles and Correlations with Gamma Radiation Dose and Precursor Concentration
Full text linkThe purpose of this research is to examine how the electro-optical behavior of platinum (Pt) nanoparticles prepared via the gamma radiolysis process is related to both the radiation dose and to the Pt precursor concentration. The Pt precursor used in these experiments has been radiolytically degraded using a <sup>60</sup>Co gamma source at dosages ranging from 80 kGy to 120 kGy. As well, varying the concentration of the Pt precursor from 5.0 &times; 10<sup>&minus;4</sup> M to 20.0 &times; 10<sup>&minus;4</sup> M was carried out as a systematic investigation. Spectrophotometric analysis utilizing UV&ndash;Visible spectroscopy and TEM provided the optical data and particle size information for the nanoparticles. The results indicate that increasing the radiation dosage results in smaller Pt nanoparticle sizes due to an increased rate of nucleation and that increasing the Pt precursor concentration leads to larger Pt nanoparticles due to an increase in ion recombination. Both the dose and concentration dependency of the optical absorption spectrum indicate a significant relationship between size and plasmon behavior. Also, the conduction band energy level, which was determined from the maximum of the UV&ndash;Visible absorption peak, is dependent on the particle size and shows a pronounced quantum confinement effect, with the conduction band energy increasing as the particle size decreases. Thus, these studies provide a definitive correlation of structure&ndash;property in Pt nanoparticles and confirm the capability of the gamma radiolytic synthesis process to be used for controlling the specific electronic and optical properties of Pt nanoparticles
CX3CR1-Fractalkine Dysregulation Affects Retinal GFAP Expression, Inflammatory Gene Induction, and LPS Response in a Mouse Model of Hypoxic Retinopathy
Full text linkDiabetic retinopathy (DR) causes vision loss due to sustained inflammation and vascular damage. The vascular damage is evident by fibrinogen leakage, angiogenesis, and hypoxia. Neuronal regulation of microglia via the CX3CL1 (Fractalkine or FKN)-CX3CR1 pathway plays a significant role in retinal pathology. Defects in FKN or CX3CR1 exacerbate inflammation, vascular damage, and vision impairment. However, the contribution of hypoxic astrocytes to the pathological process of DR is unclear. A hypoxic model (7 days of systemic 7.5% O<sub>2</sub>) was utilized to induce retinal damage in adult mice in the absence of systemic inflammatory signals. This model induced vascular and microglial responses similar to 10 weeks of STZ-induced hyperglycemia. The goal of this study is to characterize retinal damage in WT and mice with defects in the FKN-CX3CR1 signaling axis and hence assess the impact of the microglial inflammatory responses to hypoxic retinopathy. Tissues were analyzed by immunostaining, RNA sequencing, and cytokine quantification. We found that CX3CR1 deficiency in hypoxic animals induced reactive astrogliosis and that M&uuml;ller glial responses to hypoxia and systemic inflammation were dependent on FKN signaling. Exacerbated microglial reactivity to hypoxic conditions significantly altered the expression of HIF transcripts. Microglial dysregulation was found to reduce the anti-inflammatory response to hypoxic conditions, downregulate hypoxia-responsive gene expression, and restrained LPS-induced inflammatory responses. We found that microglia dysregulation alters the hypoxic response by inhibiting the upregulation of HIF2&alpha;/3&alpha;, increasing CD31 immunoreactivity, and altering the expression of ECM-associated transcripts such as type I, III, and XVIII collagens to hypoxic conditions.Molecular Microbiology and ImmunologySouth Texas Center for Emerging Infectious Disease