Marietta College

Identifying top-performing healthcare providers within healthcare systems is critical for enhancing care quality, promoting coordination, and informing data-driven health policy. Current evaluation systems, such as the Merit-based Incentive Payment System (MIPS), rely on extensive, provider-submitted data and complex composite scoring, making them difficult to compute, particularly for large healthcare systems. In this study, we present a network-based sheaf-theoretic approach for identifying top providers within healthcare systems. We first represent healthcare systems as patient-sharing provider networks, where edges encode collaborative relationships derived from shared beneficiaries. To capture how clinical expertise and collaboration flow through the healthcare network, we introduce a mathematical framework based on sheaf theory that accounts for provider specialties and the structure of their shared patient connections. Next, we use this framework to generate rankings of healthcare providers that reflect both the extent of their collaboration and the relevance of their clinical expertise. We evaluate our method on the Centers for Medicare & Medicaid Services (CMS) Physician Shared Patient Patterns dataset, enriched with provider specialty information from the National Plan and Provider Enumeration System (NPPES). We use publicly available MIPS scores as ground truth for validation. Experimental results demonstrate that our sheaf-based model performs well, and is highly effective in identifying top-performing healthcare providers.

Social identity in adolescence refers to the way young people define themselves in relation to the social groups they belong to, such as their peers, family, cultural or ethnic communities, gender, and sometimes subcultures. All of these areas of a person’s identity are also influenced by the different contexts and the social discourses and expectations that are experienced. Social and emotional skills, competencies, or capacities are internal assets that can be learned and are linked to prosocial behavior and positive social identities. Developing strong social and emotional capabilities is particularly pertinent for young adolescents due to the convergence of developmental effects as they begin to form their sense of personal identity and shift to having a greater sense of independence. This paper presents the findings from a systematic literature review using PRISMA reporting guidelines of studies conducted between 2014 and 2024 that examined the links between social and emotional competencies and a young adolescent’s social identity formation. Four education-focused databases were searched for peer-reviewed articles published in English. Results were mapped against CASEL’s social and emotional skill sets. Very limited literature was found (n = 6) that explicitly addressed young adolescent identity formation and the need for social and emotional skills; however, key findings show that young adolescents utilize all five of CASEL’s social and emotional skill sets to support the development of their social identity. Bronfenbrenner’s bioecological theory is applied to the findings to consider the systemic influences. Questions for future research are also proposed.

Context Immersive clinical education experiences (ImCEs) are a recent addition to the Commission on Accreditation of Athletic Training Education standards. As such, there is little information on how athletic training programs design and implement ImCEs into the curriculum. Objective The purpose of this study was to explore the structure of ImCEs among athletic training programs and practices relating to identifying and developing ImCEs. Design Cross-sectional study. Setting Web-based survey. Patients or Other Participants A total of 103 of 265 Coordinators of Clinical Education for Commission on Accreditation of Athletic Training Education-accredited professional programs participated (women = 69, men = 29, 4 = prefer not to disclose, 1 = unanswered). Main Outcome Measures Coordinators of Clinical Education provided information about their program, timing and length of ImCEs, and the settings used. Program practices for preceptor selection and development, curricular design for simultaneous didactic coursework, and resources available to students were also investigated. Results The average number of ImCEs was 1.9, with a length of 4 to 28 weeks. Most programs have the first ImCE in the second year and primarily rely on college/university and secondary school settings. Programs reporting ImCEs less than 4 weeks in length and those requiring synchronous coursework during clinical immersion are of concern. Conclusions Athletic training programs are integrating ImCEs in a variety of ways. There may be confusion as to best practices and Commission on Accreditation of Athletic Training Education requirements for ImCEs.

Abstract The objective of this study is to perform an in-depth assessment of the In-house Data Integration Platform (IDIP). This solution, crafted by local engineers, is tailored to fit the purpose of enhancing digitalization endeavors within the company. Notably, the platform's distinctiveness lies in its role as both a culmination of local engineering ingenuity and a valuable input source for various commercial software applications. The IDIP employs a rigorous multi-disciplinary approach, integrating in-house programming, customized developed algorithms, and statistical methods alongside machine learning technologies. The evaluation methodology focuses on systematically centralizing, collecting, and structuring data from diverse sources. This includes a spectrum from Excel files and SQL databases to extensive big data, such as videos, hard files etc., spanning various departments. This is coupled with the identification of shortcomings in existing commercial tools to cover. The ultimate objective is to efficiently address the daily operational requirements of the company's technical team, minimizing efforts and time. The introduction of the locally developed platform has led to significant results, promoting increased teamwork and integration among various teams. The platform's impact is evident in its efficient handling and reporting of oil and gas reserves for over 400 fields with more than 500 different reservoirs. This achievement has substantially reduced completion time from over a month to about an hour. Additionally, the IDIP offers extra advantages, such as accommodating up to 500 simultaneous users engaging in different tasks with distinct input systems. This capability caters to a diverse user base and ensures smooth operations. In terms of production oversight, the platform serves both technical and non-technical users, providing a quick identification of any discrepancies in production allocation. The structured data provided by the platform is beneficial for data science and large language models, enabling easy utilization and contributing to efficient and organized data handling. The platform also establishes a real-time feedback loop, allowing swift reporting of suggestions or improvements. The IDIP not only improves multidisciplinary workflows but also plays a crucial role in preserving and capturing local engineering ingenuity. This study highlights its pivotal role in shaping a shared vision of subsurface scenarios. A distinctive feature of the platform lies in its seamless integration of machine-learning algorithms and data-driven models, providing a novel, flexible, and locally crafted contribution to the digitalization landscape. Additionally, the platform leverages the ingenuity of local engineers by depending on open-source libraries and frameworks, allowing for easy customization without the need for cumbersome workarounds.

In recent years, machine learning (ML) has evolved from a niche field to a transformative technology that empowers businesses to make data-driven decisions, automate tasks, and enhance customer experiences. As machine learning (ML) continues to redefine the technological landscape, the transition from conceptualizing ML models to operationalizing them is at a critical juncture. MLOps, the amalgamation of practices and tools facilitating the seamless integration of ML system development and operations, emerges as the linchpin in this transformative journey.