Search Results (343)

Non-intrusive load monitoring (NILM), as a key technology in smart-grid advanced metering infrastructure, aims to disaggregate mains power from smart meters into individual load-level power consumption. Traditional NILM methods require centralizing sensitive measurement data from users, which poses significant privacy risks. Federated learning (FL) enables collaborative training without centralized measurement data, effectively preserving privacy. However, FL-based NILM systems face serious threats from attacks such as model inversion and parameter poisoning, and rely heavily on the availability of a central server, whose failure may compromise measurement robustness. This paper proposes a hybrid FL framework that dynamically switches between centralized FL (CFL) and decentralized FL (DFL) modes, enhancing measurement privacy and system robustness simultaneously. In CFL mode, layer-sensitive pruning and robust parameter aggregation methods are developed to defend against model inversion and parameter poisoning attacks; even with 30% malicious clients, the proposed defense limits the increases in key error metrics to under 15.4%. In DFL mode, a graph attention network (GAT)-based dynamic topology adapts to mitigate topology poisoning attacks, achieving an approximately 17.2% reduction in MAE after an attack and rapidly restoring model performance. Extensive evaluations using public datasets demonstrate that the proposed framework significantly enhances the robustness of smart-grid measurements and effectively safeguards measurement privacy. Full article

Background: Effective communication is a fundamental competency in veterinary medicine that shapes the quality of veterinarian–client relationships, shared decision-making, and animal welfare. However, consistent and systematic integration of communication training across veterinary curricula remains uneven worldwide. Methods: This scoping review mapped and analyzed educational programs aimed at developing communication competencies in veterinary education and professional practices. A systematic search was conducted according to PRISMA-ScR guidelines, identifying 37 eligible studies published between 2005 and 2024. Results: Most publications were in English and originated from North America, particularly Canada (n = 15) and the United States (n = 8). Regarding target populations, 15 studies (40.5%) focused on veterinary students, 12 (32.4%) on practicing veterinarians, 8 (21.6%) on animal owners or clients, and 2 on veterinary educators. 18 studies (48.7%) described structured programs that used active learning strategies such as role-play, clinical simulations, peer-assisted learning, and formative feedback. The competencies frequently emphasized include empathy, active listening, nonverbal communication, conflict resolution, and rapport building. Notable best practices included the Calgary–Cambridge model, Objective Structured Clinical Examination (OSCE), and reflective video analysis. Conclusions: The available evidence indicates a growing emphasis on clinical communication within veterinary education, primarily implemented through experiential and practice-based approaches. However, substantial gaps persist in the representation of Latin American contexts and in the systematic, longitudinal integration of communication skills across veterinary curricula. Addressing these gaps may contribute to more coherent, equitable, and context-sensitive communication training in veterinary education. Full article

As encrypted traffic analysis becomes increasingly vital for network security, the conventional reliance on centralized classification faces growing challenges due to data privacy regulations and data silos across heterogeneous nodes. Federated learning (FL) emerges as a solution by training models locally and sharing only parameter updates, thus preserving privacy. However, its performance is significantly degraded by data heterogeneity (i.e., non-IID data) among participants. To address this critical challenge, this paper proposes a Federated Learning framework based on Deep Mutual Learning (FLDML). In this method, clients first train local models on their private traffic data and then upload them to a server. There, they engage in deep mutual learning through co-training on a shared public dataset to enhance robustness and mitigate data heterogeneity. Subsequently, a global classifier is generated by averaging the model parameters. When evaluated on the ISCX VPN-NonVPN 2016 dataset, FLDML demonstrates significantly superior performance in handling non-IID traffic data compared to classical FL algorithms. This study concludes that the proposed framework not only effectively mitigates data heterogeneity in federated scenarios but also provides a scalable and improved solution for distributed network traffic classification. Full article

Background: Human trafficking affects millions of people worldwide with multiple adverse outcomes including psychopathology. Although research on human trafficking has become abundant in other academic disciplines (e.g., public health, criminology, social work), healthcare research specific to the mental health treatment of survivors remains limited. Objective: The purpose of this study was to gather recommendations from professionals about mental health treatment of trafficking survivors. Method: Semi-structured interviews were conducted with 21 multidisciplinary professionals working with trafficking survivors. Data were analyzed using qualitative content analysis methodology. Results: An overarching theme derived from the data concerned the recommendation to implement trafficking survivor-informed care, specifically addressing complex trauma and individual client contexts, such as culture and history prior to trafficking. A second theme emphasized the recommendation for comprehensive care, achieved through outreach efforts, interdisciplinary services, case management, and ongoing training for mental health professionals. Conclusions: Professionals working with trafficking survivors perceived conventional service formats as insufficient, and they recommended personalized and comprehensive healthcare to address multiple needs and extensive trauma history. Full article

Background/Objectives: Integrating behavioral health and primary care services is a national public health priority in the US, especially in underserved settings like northern Arizona. This healthcare delivery model is crucial to meet the mental and physical health needs of people with SU/SUDs, particularly those belonging to culturally diverse populations. In collaboration with a behavioral health center in northern Arizona, the current study aimed to assess the perspectives of providers and administrative staff on the implementation of integrated primary care (IPC) services for people with SU/SUDs. Methods: In February 2023, twelve healthcare providers and administrative staff from diverse educational backgrounds were recruited using purposive sampling to capture a range of perspectives on IPC implementation at the behavioral health center. Participants completed individual, semi-structured interviews conducted via Zoom, which were audio recorded and lasted approximately 30 min. The interview recordings were transcribed verbatim using Trint Software, and analyzed on Google Docs using applied thematic analysis. Two researchers coded the transcripts, iteratively developing and refining themes through multiple cycles of review and team discussions. Additional team members provided feedback and verified the themes, with consensus reached through collaborative meetings. This rigorous, iterative approach ensured the reliability and validity of the final thematic framework. Results: We found that IPC supports SU/SUDs recovery by providing holistic care that integrates medical, mental health, and addiction services while addressing social and co-occurring needs. It fosters an empathetic environment where clients do not need to repeatedly disclose their SU/SUDs, improves access to preventive care, and offers continuous support and education. Implementation barriers included workforce shortages, limited internal communication, and insufficient interdisciplinary training. Gaps in culturally centered care were identified, including reliance on Western models, limited representation of Native American and sexual and gender minority staff, and inconsistent use of inclusive practices such as pronouns, traditional healing, and trauma-informed approaches. Additionally, community partnerships with multisectoral organizations help clients access supportive resources beyond the facility, including vision care, clothing, and dental services. Conclusions: The implementation of IPC was seen as important to support the behavioral health center in northern Arizona to foster an empathetic environment where clients with SU/SUDs can have their mental, physical, and social needs addressed, either within the facility or through community partnerships, thereby supporting their recovery. However, progress is hindered by barriers such as workforce shortages, limited internal communication, and insufficient interdisciplinary care training. Additionally, despite regular cultural competency training, gaps remain in culturally centered care for underserved populations, particularly Native American and sexual and gender minority clients. Full article

Federated representation learning (FRL) is a promising technique for learning shared data representations that capture general features across decentralized clients without sharing raw data. However, there is a risk of sensitive information leakage from learned representations. The conventional differential privacy (DP) mechanism protects the privacy of the whole data by randomizing (adding noise or random response) at the cost of deteriorating learning performance. Inspired by the fact that some data information may be public or non-private and only sensitive information (e.g., race) should be protected, we investigate the information-theoretic protection on specific sensitive information for FRL. To characterize the trade-off between utility and sensitive information leakage, we adopt mutual information-based metrics to measure utility and sensitive information leakage, and propose a method that maximizes the utility performance, while restricting sensitive information leakage less than any positive value $ϵ$ via the local DP mechanism. Simulation demonstrates that our scheme can achieve the best utility–leakage trade-off among baseline schemes, and more importantly can adjust the trade-off between leakage and utility by controlling the noise level in local DP. Full article

Drawing on organizational justice theory, this study examines how workers’ perceptions of fair treatment influence job satisfaction following the transition from temporary agency employment at subcontracting firms to regular employment with client firms. A multilevel analysis was conducted to simultaneously assess individual- and organizational-level effects. Data were collected through a survey of Korean public organizations that had implemented regular employment transitions, yielding a final sample of 966 employees nested within 116 institutions. At the individual level, perceived fair treatment after regularization was positively associated with job satisfaction. At the organizational level, systematic human resource management practices enhanced employees’ perceptions of fair treatment, while government support during the transition process—including the provision of clear guidelines and professional consulting—moderated the relationship between perceived fair treatment and job satisfaction. These findings contribute to a deeper understanding of how fair treatment perceptions shape employee attitudes following employment regularization and highlight the role of organizational human resource practices and government involvement in fostering positive outcomes during labor market transitions. Full article

Finger vein recognition algorithms based on deep learning techniques are widely used in many fields. However, the training of finger vein recognition models is hindered by privacy issues and the scarcity of public datasets. Although applying federated learning techniques to finger vein recognition can effectively address privacy concerns, data heterogeneity across clients limits the performance of the models, especially on small datasets. To address these problems, in this paper, we propose a new federated finger vein recognition algorithm (FedPSFV). The algorithm is based on the federated learning framework, which increases the interclass distance of each dataset by sharing the prototypes among clients to solve the data heterogeneity problem. The algorithm also integrates and improves the margin-based loss function, which advances the feature differentiation ability of the model. Comparative experiments based on six public datasets (SDUMLA, MMCBNU, USM, UTFVP, VERA, and NUPT) show that FedPSFV has better accuracy and generalizability; the TAR@FAR = 0.01 is improved by 5.00–11.25%, and the EER is reduced by 81.48–90.22% compared to the existing methods. Full article

Because auditors have access to corporate information, a firm’s decision to protect material trade secrets should, in principle, not influence audit effort. We analyze the effects of trade secrecy protection on the audit fees, documenting that firms with redacted information pay significantly higher fees than those that do not redact information. In cross-sectional tests, we further document that the relationship between redaction and audit fees is significantly influenced by both auditor and client characteristics. Consistent with the premise that redaction increases the complexity of the audit—particularly if auditors view redacted disclosures as indicators of potential business or litigation risk—the regression results indicate that the main effect is moderated by auditor factors such as specialization, tenure, and quality, as well as client factors like existing relationships, bargaining power, and reporting quality. These insights contribute to ongoing discussions in audit policy by illustrating how confidential disclosure practices affect audit effort and costs. Overall, our results inform policymakers seeking to reconcile firms’ proprietary information protection with public interest in transparent and credible financial reporting. Full article

Food insecurity is a pervasive public health issue in the United States. While food pantries attempt to alleviate this issue, their effectiveness is limited by structural and logistical barriers that affect service accessibility. Transportation is a frequently underexamined barrier for individuals trying to access food aid. The purpose of this study is to assess the interplay of client- and pantry-level characteristics and their influence on food aid accessibility across several transportation modalities. This cross-sectional survey study collected data from 430 food pantry clients concerning their demographics, transportation methods, and perceptions of transportation barriers. Pantry characteristics were also collected focusing on transportation infrastructure and operational policies. Individual and grouped comparisons were made between transportation methods in relation to pantry visitation, with those walking, biking, and taking a bus to the pantry grouped to compare to those taking a car. Higher food insecurity score, smaller household size, single relationship status, and race were independently associated with increased odds of walking, biking, or taking a bus to the pantry. Having closer bus stops, more bus lines, and no monthly use limits were independently associated with increased odds of walking, biking, or taking a bus to the pantry. Several characteristics were associated with specific transportation modalities when accessing food aid. Our results are particularly concerning given the increased food insecurity and additional vulnerabilities seen in those who walk, bike, or take the bus to the pantry. Transportation disadvantage may be ameliorated by less restrictive pantry use policies and more robust public transit. Full article

In the 21st century, community participation in heritage management has become a worldwide phenomenon. Despite this shift in heritage management thinking and the inclusion of community members in such processes, these participatory efforts have not necessarily led to the empowerment of communities in decision making. Many studies have been conducted to probe this critical question and to better understand what hinders the active involvement of the public. This paper examines the question through Crossing the Pell, an academic project from the Rhode Island School of Design (RISD). With grant sponsors/clients and based in real-time heritage management issues of historic Newport, Rhode Island, this project emphasized creative and untested methods for engaging community through the use of immersive digital environments. As sponsored design with the stipulated objectives of funders and clients, the professors and students were charged with not only solutions for an adaptation of existing infrastructure but also the design of a unique community engagement process. This paper documents the multi-year academic endeavor, analyzes its outcomes in the context of the history of community engagement, and offers a hypothesis for implementing an active and meaningful participatory process in cultural heritage management through the use of extended realities. Full article

The integration of Internet of Things (IoT) technologies into public healthcare enables continuous monitoring and sustainable health management. However, conventional frameworks often depend on transmitting and storing raw personal data on centralized servers, posing challenges related to privacy, security, ethical compliance, and long-term sustainability. This study proposes a privacy-preserving framework that avoids the exposure of true health-related data. Sensor nodes encrypt collected measurements and collaborate with a secure computation core to evaluate health indicators under homomorphic encryption, maintaining confidentiality. For example, the system can determine whether a patient’s heart rate within a monitoring window falls inside clinically recommended thresholds, while the framework remains general enough to support a wide range of encrypted computations. A compliance verification client generates zero-knowledge range proofs, allowing external parties to verify whether health indicators meet predefined conditions without accessing actual values. Simulation results confirm the correctness of encrypted computation, controllability of threshold-based compliance judgments, and resistance to inference attacks. The proposed framework provides a practical solution for secure, auditable, and sustainable real-time health assessment in IoT-enabled public healthcare systems. Full article

Introduction: Oral health problems are prevalent among people receiving palliative care, affecting their quality of life. However, little is known about dental practitioners’ perspectives in this setting. Thus, this study aimed to explore the perceptions of public dental practitioners regarding the provision of dental care for people who receive palliative care. Methods: An exploratory focus group with 21 public dental practitioners, with a mean of 8.24 years of experience, was conducted in a public oral health service in Sydney, Australia. The focus group was transcribed and analysed using thematic analysis. Results: Three themes were identified: (1) enhancing quality of life in palliative care through improved oral health; (2) navigating the systemic and practical challenges of palliative dental care; (3) competent, collaborative, and optimised: a palliative oral care model. Participants highlighted the importance of oral health to quality of life while receiving palliative care. Nevertheless, there were several systematic and practical challenges to delivering appropriate dental care, which included competing priorities among clients, disconnects and gaps in care coordination with palliative care providers, limited training, and adapting treatment planning during appointments. Participants highlighted the need for a new model of care in the future that improved dental practitioners’ competence through comprehensive palliative dental training, included protocols to facilitate interdisciplinary collaboration, and optimised dental treatment planning and appointment scheduling. Conclusions: Public dental practitioners in this study demonstrated positive attitudes, but systemic barriers and limited training restrict their care provision. A palliative oral health care model for this setting should include palliative dental training, foster interdisciplinary collaboration, and optimise dental treatment planning. Full article

Lyme borreliosis (LB), a zoonotic disease caused by Borrelia burgdorferi sensu lato and transmitted by Ixodes ricinus, is increasingly relevant in France, yet data on public awareness and prevention are limited. This cross-sectional study assessed the knowledge, attitudes, and practices (KAP) of companion-animal owners (CAOs) and veterinarians regarding LB. Online questionnaires were distributed between January and April 2023, yielding 159 valid CAO responses and 48 responses from veterinarians. Among owners, 95% reported outdoor access for their pets, 73.9% had observed ticks, and 88.8% used ectoparasiticides. However, only 16.4% reported LB vaccination, and just 38% recognized key canine clinical signs. Veterinarians frequently suspected LB (70%) but confirmed diagnoses less often (39.6%), relying mainly on clinical signs and the results of rapid tests. While prevention through antiparasitic use was common, communication between veterinarians and owners regarding LB was inconsistent. Despite this, most owners reported compliance with veterinary guidance. These findings reveal good preventive habits but identify knowledge gaps and underdiagnosis. Enhancing veterinary–client communication and promoting a One Health approach could improve LB prevention and control in France. Full article

Purpose: This article evaluated the decline of registered auditors (RAs) and its impact on the future of the assurance industry in South Africa. Auditors play a critical role in ensuring the transparency, trust, and credibility of financial statements. The decrease in the number of registered auditors has become a pressing issue, raising concerns about the assurance industry’s ability to maintain a sufficient number of registered auditors and continue providing assurance services to public and private entities. Methodology: A qualitative Delphi methodology was employed, involving interviews with RAs who are registered with the Independent Regulatory Board for Auditors (IRBA). Eight RAs participated in structured interviews. This approach enabled the researcher to gather expert opinions, identify emerging trends, and explore challenges and opportunities within the audit profession related to the decline of RAs. Main findings: The decline of RAs is straining client demands, increasing workloads, and leading to a shortage of audit firms, which in turn affects audit quality and methodologies. Audit firms struggle to attract and retain talent due to regulatory burdens, economic pressures, and concerns about work–life balance. These pressures have resulted in higher audit fees, increased compliance costs, and more extensive training requirements. Smaller audit firms are especially impacted, with some downscaling their assurance services or exiting the market entirely. Practical implications: This study underscores the pressing need for regulatory bodies, such as the IRBA, to address the challenges faced by audit firms, particularly in terms of compliance and workforce retention. Proactive strategies are required to preserve the quality and accessibility of assurance services. Contribution: This study contributes to the ongoing discourse on the future of the audit profession by offering grounded insights into how the industry might sustain itself amid a declining number of RAs and changing professional dynamics. Full article