Bowie State University

Intelligent human motion analysis is essential for developing next-generation IoT and AR/VR systems that enable automated, interpretable, and fine-grained performance assessment. Motivated by the need for real-time, explainable, and transferable skill evaluation, we propose a wearable sensing framework to assess human performance by tracking skill progression and minimizing injury risk. We use live badminton gameplay and workout exercises as representative use cases, where motion dynamics, postural stability, and limb coordination are critical to success. Both activities demand optimal posture and synchronized limb movements, while improper actions or suboptimal technique can lead to decreased performance and higher injury susceptibility. We introduce SkillNet , a multi-task learning framework that extracts shared representations across all limbs while preserving limb-specific motion signatures. The architecture employs task-specific regressors to detect subtle inter-limb dissimilarities and distinctive traits, enabling collective inference in a body sensor network (BSN) environment. To holistically measure performance, we formulated a weighted performance indicator (PI) that fuses AI-driven scoring with domain-expert evaluations, providing a robust metric for both qualitative and quantitative assessment. We evaluate SkillNet on three diverse datasets B adminton A ctivity R ecognition (BAR), M ulti- M odalities D ataset of S ports (MMDOS), and D aily and S ports A ctivities (DSADS) capturing a broad spectrum of motion types and skill intensities. Results show that SkillNet achieves an R \({}^{2}\) score of 86% and a mean squared error of 0.0093 in performance prediction. The integrated AI–expert scoring mechanism improves baseline performance estimation by 14.95% , demonstrating the advantage of combining human expertise with automated analysis. We further benchmark inference time, memory usage, and power consumption of the SkillNet , validating its efficiency and feasibility for real-time, end-to-end task inference on resource-constrained embedded edge devices, Jetson Nano and Jetson Xavier NX platforms.

Abstract There are many pressing scientific questions surrounding the topic of whether forest management has resulted in higher burn severity from recent wildfires in timberlands of the Pacific states. Using burn severity maps from Landsat satellite imagery, zonal statistics in QGIS were used to summarize and compare the attributes (mean, median, variance, range) of burn severity classes within two zones for each fire: the privately managed forest area and a surrounding control (largely unmanaged forested area). We analyzed 100 individual managed forest areas across the Pacific states with a total of 800 privately owned management units. Comparison of the burned severity class by individual managed forest area showed that 42% of these timberlands burned at significantly lower severity ( p < 0.05) than their surrounding (unmanaged) buffer zones in large wildfires between 2013 and 2022. In addition, 30% of managed forest lands were not significantly different from their unmanaged buffer zones in burn severity. Landsat normalized difference moisture index (NDMI) clearly shows recent clear-cuts, fire scars, and thinning management in every case we examined, eliminating the possibility of underestimating or overlooking timber management activities in control buffer zones. The highest burn severity did occur in landscapes where extremely high levels of pre-fire live forest biomass remained in large patches around equally large and thinned or logged forest areas. We conclude that this type of mixed-age management plan may create a potentially explosive fuel-loading status in a forest, whereby wildfire can be readily carried by high winds from dense fuel areas (not recently thinned or managed) over and around patches of low biomass stands that have been recently thinned and logged. NB: References will be numbered and cited in order upon acceptance of the paper

Abstract Endophytic fungi that asymptomatically reside in plant tissues are of growing interest as promising sources of biologically active agents. In this study we identified endophytic fungi from healthy flowers, bark, leaves, fruit and resin of the traditionally used medicinal plants Azadirachta indica A. Juss (neem) and Melia azedarach L (melia) in Kenya. . A total of 95 fungi were isolated from neem and 46 from melia. More fungi were isolated from leaves of neem (80%) and melia (70%), compared with bark, fruits and resins. Asomycetous fungi were the most commonly isolated fungi from the two plant species, including Phomopsis sp., Penicillium sp., Colletotrichum sp. and Preussia sp. Neem exhibited high fungal richness compared with melia. Geographic differences in predominant species were observed. Characterizing endophytic fungi in medicinal plants has the potential to identify producers of secondary metabolites that may be of medicinal interest, and studying these fungi in areas of Africa where surveys have not yet been conducted opens the door to a new understanding of endophyte biodiversity.

ABSTRACT This conceptual article focuses on the Ujima Center located at Bowie State University. It illustrates Bowie State University's strong legacy of training counselors in evidence‐based multicultural practices and the Ujima Center's innovative framework that promotes diversity and equity in school counseling through culturally affirming training. The Ujima Center developed four models;AFFIRM, SAFE, HEART, and GUIDE,—to provide culturally affirming training, advisement, support, and career assistance to Ujima scholars. In addition, the Ujima Center highlights how additional evidence‐based components, such as multitiered systems of support, as well as mental health first aid training, help address the significant inequities in the availability of mental health services for students in K‐12 schools. This conceptual article also explores implications for counselor education, research, and practice.

The increasing global adoption of electronic tax systems inherently introduces significant privacy and security risks, primarily stemming from the reliance on cloud infrastructure for storing and processing highly sensitive financial data. Conventional digital tax platforms typically necessitate unrestricted access to taxpayers’ raw data, thereby rendering these systems acutely vulnerable to sophisticated cyberattacks, large-scale data breaches, and malicious insider threats. This exposure fundamentally compromises the confidentiality of personal financial records and demonstrably contributes to the erosion of public trust in governmental digital services. To address these challenges, we introduce a privacy-preserving framework specifically engineered for secure tax calculation. Our technical solution is founded on the strategic integration of Fully Homomorphic Encryption (FHE), specifically employing the Cheon-Kim-Kim-Song (CKKS) scheme. The CKKS scheme is uniquely suited to enabling approximate arithmetic on encrypted data, which facilitates the secure evaluation of complex, real-valued inputs, including income figures, allowable deductions, and financial risk metrics. We implemented an encrypted tax pipeline utilizing the CKKS scheme. This pipeline rigorously supports the necessary real-valued operations and ensures the secure computation of core tax outcomes, including the exact tax owed, potential refund amounts, and predictive fraud assessment, with inherent implications for compliant auditing and maintaining evidentiary integrity. Experimental results conclusively demonstrate that our proposed system maintains both high utility and accuracy in its calculations while simultaneously guaranteeing data confidentiality. This approach establishes a practical foundation for building secure, transparent, and trustworthy digital tax infrastructures.

<h3>Objectives</h3> To characterize treatment patterns and healthcare resource utilization in patients with idiopathic pulmonary fibrosis initiating antifibrotic therapy. <h3>Methods</h3> A retrospective cohort study was conducted using the 100% Medicare Fee-For-Service data to identify patients who initiated antifibrotic therapy (nintendanib or pirfenidone) between 01/01/17 and 12/31/22. Eligible patients were aged ≥18, had ≥1 inpatient or ≥2 outpatient IPF diagnosis claims pre-index, continuously enrolled for 6-months prior and had no prior antifibrotic therapy or lung transplant. Discontinuation, using a 45-day gap definition, per-person per-month (PPPM) healthcare resource utilization and mortality were examined. Patients were censored at the earliest occurrence of switching to the other antifibrotic, lung transplant, death, end of enrollment, or end of data. Unadjusted time-to-event was estimated using Kaplan-Meier Cox-proportional hazards model identified characteristics associated with discontinuation. <h3>Results</h3> 5,397 patients met selection criteria, 3,555 initiating nintendanib and/or 2,496 initiating pirfenidone. For all patients, mean (SD) age at initiation was 75.6(7.3), with 38% female, 87% white race and mean (SD) Deyo Charlson comorbidity index score of 2.6(2.4). 25% of patients discontinued antifibrotic therapy at 3 months and 50% at 11 months. Mean (SD) PPPM inpatient hospitalizations increased from 0.07(0.13) in the pre-index to 0.10(0.26) in the post-index. 25% and 50% of patients died at 18 and 41 months after initiation, respectively. Patients that were aged ≥75 were more likely to discontinue (HR 1.23; 95%CI: 1.14–1.33) therapy relative to those aged &lt;75. Men (HR:0.76; 95%CI:0.71–0.81) and patients that experienced a hospitalization prior to initiation (HR:0.85; 95%CI: 0.78–0.94) were less likely to discontinue therapy, after adjustment. <h3>Conclusions</h3> Patients with IPF have high treatment discontinuation to antifibrotic therapy, hospitalizations and mortality. A large unmet need remains in patients with idiopathic pulmonary fibrosis.

Cloud computing has become a cornerstone of modern IT infrastructure, offering scalability and efficiency but also exposing organizations to evolving cyber threats such as data breaches, insider threats, and advanced persistent threats (APTs). Traditional security mechanisms struggle to address these dynamic challenges, necessitating the integration of AI-driven threat detection and prevention strategies. This conceptual paper explores the comparative effectiveness of supervised learning, unsupervised learning, reinforcement learning, and hybrid AI models in cloud security. Supervised learning excels in identifying known attack patterns, while unsupervised learning is crucial for detecting zero-day threats and anomalies. Reinforcement learning enables self-adaptive security measures, and hybrid models offer a comprehensive, multi-layered approach to cloud security. However, AI-driven cybersecurity faces significant challenges, including data privacy risks, bias in threat detection, adversarial AI attacks, and lack of model interpretability. Emerging AI trends such as federated learning, quantum security, and explainable AI (XAI) are shaping the future of cloud security, while regulatory frameworks like GDPR, NIST AI Risk Management, and the EU AI Act play a crucial role in standardizing ethical AI use. This study provides insights into the strengths, weaknesses, and future directions of AI-driven cloud security, offering recommendations for researchers, policymakers, and cybersecurity practitioners to enhance AI resilience against emerging threats.

Some giant US conglomerates are now undergoing corporate spin-offs or are considering such spin-offs in the near future. Corporate spin-offs offer a unique opportunity to assess corporate capital structure decisions. The leverage ratio of the spin-off firms represents their initial capital structure. We investigate the capital structure of corporate spin-offs and find evidence that they adhere to the trade-off theory. This study provides evidence that the subsidiary firms tend to aim for a target capital ratio during the sample period. The results indicate that the partial adjustment model with firm fixed effects is a good fit for the data sample. The parent companies in corporate spin-offs exhibit a similar pattern but with a slower adjustment speed. The tendency to target capital ratios is observable in both market value and book value leverage measures for the parent and subsidiary firms. Indicators of the pecking order assumption do not possess statistically significant coefficients. Changes in share price affect market debt ratios in the short term. With alternative definitions of leverage, the estimated adjustment speeds vary. In the case of longer horizons, the results align with a continuous rate of adjustment.