Client Participation Research Articles

Provision of Digital Primary Health Care Services: Overview of Reviews.

Digital health is a growing field, and many digital interventions have been implemented on a large scale since the COVID-19 pandemic, mainly in primary health care (PHC). The development of digital health interventions and their application in PHC are encouraged by the World Health Organization. The increased number of published scientific papers on this topic has resulted in an overwhelming amount of information, but there is no overview of reviews to summarize this evidence. This study aims to provide policy makers, health managers, and researchers with a summary of evidence on digital interventions used in PHC. This overview of reviews searched the Web of Science and MEDLINE databases for systematic and scoping reviews on assessments of digital technologies implemented in PHC published from January 2007 to March 2023. Only reviews that addressed digital interventions whose targets were real patients or health care providers (HCPs) were included. A total of 236 records were identified from the search strategy, of which 42 (17.8%) full-text papers were selected for analysis, and 18 (7.6%) reviews met the eligibility criteria. In total, 61% (11/18) of the reviews focused their analysis on specific digital health interventions (client-to-provider telemedicine, provider-to-provider telemedicine, health worker decision support systems, systems for tracking patients' health status, client participation and self-care platforms, and provision of education and training to health workers), and 39% (7/18) of the reviews focused on specific topics related to PHC (preventive care, chronic disease management, behavioral health disorders, the COVID-19 pandemic, multicomponent PHC interventions, and care coordination). Most studies in the included reviews agreed on barriers to implementation, such as software and apps developed without involving end users, the lack of training of HCPs and patients in digital technology use, and the lack of reimbursement and billing strategies for remote consultations. However, they showed several mixed results related to health service quality and patients' clinical conditions and behavior changes. Research in digital health applied to PHC is still concentrated in high-income countries, mainly in North America and Europe. The mixed results related to health service quality and patients' clinical conditions or behavior changes may have been caused by deficiencies in the process of implementing digital interventions. It is necessary to examine the entire impact pathway and the causal relationship among implementation, health service quality, and clinical condition outcomes to support the spread of digital health in PHC settings.

Enhancing federated learning with dynamic weight adjustment based on particle swarm optimization

Federated learning (FL) stands as a promising distributed machine learning approach today, allowing model training on local clients without data sharing. However, varied contributions among clients often impact the global model’s performance, while FL confronts challenges like communication overhead, data heterogeneity, and privacy concerns. In this paper, we introduce a novel federated learning server aggregation algorithm: the Federated Learning Algorithm with Optimized Weight Aggregation via Particle Swarm Optimization Algorithm (AdpFedPSO). This approach dynamically adjusts client model contribution weights based on their performance and stability, aiming to enhance global model accuracy, convergence speed, and make model aggregation smarter and more adaptable. Through experimental validation on real datasets, we find that AdpFedPSO enhances accuracy by about 15% on the 0.6-Dirichlet MNIST dataset, 7.3% on the FashionMNIST dataset, and 13.4% on the CIFAR-10 dataset compared to the traditional FL aggregation method, FedAvg. These results indicate that AdpFedPSO not only enhances the accuracy of the global model but also expedites the model’s convergence speed. Additionally, it demonstrates resilience across various levels of client numbers, participation rates, and client heterogeneity, providing valuable reference and guidance for the further development of FL technology. Moreover, the concept of employing the particle swarm optimization algorithm for FL model aggregation also offers new insights and directions for future research.

Impact of family engagement on client participation in coordinated specialty care for first episodes of psychosis

BackgroundCoordinated specialty care (CSC) programs for first episodes of psychosis are increasingly offered in the United States. A component of CSC programs is active family engagement in treatment, though research on the impact of this engagement is limited. This study examined the characteristics of families engaged compared to families not engaged in treatment, and the impact of family engagement on client participation and medication adherence over the first 6 months of treatment.MethodsUsing data from the Early Psychosis Intervention Network (EPINET) research hub in Minnesota (EPI-MINN), we compared two groups of individuals: clients who had a family member(s) engaged in their treatment vs. clients who did not. Family engagement was defined as any treatment services provided to a family member(s) by CSC clinical staff. The groups were compared on intake demographic variables, duration of untreated psychosis (DUP), hospitalizations, symptom severity, and functioning. A comparison of the total number of treatment visits during the first 6 months of treatment was tested using both nonparametric (Mann Whitney U) and parametric (ANCOVA) tests. Group comparisons on self ratings of “intent to attend visits,” “intent to complete the program,” and medication adherence were tested with ANCOVA and Chi-Square.ResultsFamily-engaged clients were younger, with less years of education, and more often White; clients without family engagement were more often Black. Family engagement was positively associated with increased total number of visits for all interventions with the exceptions of client peer support and case management visits. Family engagement increased clients’ self-reported intent to attend visits, though not intention to complete the program, which was moderately to markedly high in both groups. No differences were noted with medication adherence, with high rates of adherence across the entire study sample.ConclusionsOverall, results of the study support the benefits of family engagement in CSC on client participation, though future research is needed to understand why Black families are less engaged and what treatment adaptations are needed to reduce these racial differences. The results also support the value of CSC programs for medication adherence, a critical factor in symptom reduction and mental health recovery.

Current and Future Implementation of Digitally Delivered Psychotherapies: An Exploratory Mixed-Methods Investigation of Client, Clinician, and Community Partner Perspectives.

Following the initial outbreak of the COVID-19 pandemic, mental health clinicians rapidly shifted service delivery from in-person to digital. This pivot was instrumental in maintaining continuity of care and meeting increased mental health service demands. Many mental health services have continued to be offered via digital delivery. The long-term implications of delivering services via digital media remain unclear and need to be addressed. This study aimed to identify current micro (i.e., clinician-patient interactions), meso (i.e., clinician-clinic manager interactions), and macro (i.e., government-policy maker interactions) level issues surrounding the use of digital mental health interventions (DMHI). Such integrated assessments are important for optimizing services to improve treatment outcomes and client satisfaction. Participants were recruited between January 2022 and April 2023. Quantitative data were collected using a survey informed by the Hexagon Tool. Qualitative data were collected from online semi-structured interviews and focus groups and analyzed using rapid thematic analysis. Survey data were collected from 11 client and 11 clinician participants. Twenty-six community partner participants were interviewed for this study. Client and clinician participants expressed satisfaction with the implementation of DMHI. Community partner participants generally agreed, reporting that such services will play an integral role in mental healthcare moving forward. Community partners shared that certain issues, such as uncertainty surrounding policies and regulations related to digital delivery, must be addressed in the future. Participants in this study supported the use of DMHI despite difficulties implementing these programs, asserting that such services are not a temporary fix but a pivotal cornerstone in the future of mental healthcare service delivery.

Multiple alliances and client participation during joint client meetings: an observational study

This article reports on a Dutch observational study to examine which interactions promoted multiple alliances and client participation during joint client meetings in social work and financial and employment services. It seeks to give professionals and organisations insight into how these meetings can be improved so that they deliver more for clients with multiple life issues. The central question was: ‘Which interactions (verbal and non-verbal) contribute to multiple alliances, and what place does client participation have in multiple alliances?’ An observational study with video recordings of joint client meetings (N = 20) was conducted. Analysis revealed insights into interactions, with a focus on five sensitising concepts: (a) team vision/goals; (b) shared decision making; (c) conflicts; (d) team support; and (e) open communication. Results indicate the central importance of the sequence of introducing the client, open communication, team support, team vision/goals and shared decision making in either facilitating or challenging the development of client-centred multiple alliances. Centralising the client’s voice proves difficult, and divisions between clients and the several professionals arose many times because the emotional attachment fell short. This article provides deeper insight into behaviours and interactions that are of the utmost importance and contribute to empowering multiple alliances during joint client meetings.

Stabilizing and Accelerating Federated Learning on Heterogeneous Data With Partial Client Participation.

Federated learning (FL) commonly encourages the clients to perform multiple local updates before the global aggregation, thus avoiding frequent model exchanges and relieving the communication bottleneck between the server and clients. Though empirically effective, the negative impact of multiple local updates on the stability of FL is not thoroughly studied, which may result in a globally unstable and slow convergence. Based on sensitivity analysis, we define in this paper a local-update stability index for the general FL, as measured by the maximum inter-client model discrepancy after the multiple local updates that mainly stems from the data heterogeneity. It enables to determine how much the variation of client's models with multiple local updates may influence the global model, and can also be linked with the convergence and generalization. We theoretically derive the proposed local-update stability for current state-of-the-art FL methods, providing possible insight to understanding their motivation and limitation from a new perspective of stability. For example, naively executing the parallel acceleration locally at clients would harm the local-update stability. Motivated by this, we then propose a novel accelerated yet stabilized FL algorithm (named FedANAG) based on the server- and client-level Nesterov accelerated gradient (NAG). In FedANAG, the global and local momenta are elaborately designed and alternatively updated, while the stability of local update is enhanced with help of the global momentum. We prove the convergence of FedANAG for strongly convex, general convex and non-convex settings. We then conduct evaluations on both the synthetic and real-world datasets to first validate our proposed local-update stability. The results further show that across various data heterogeneity and client participation ratios, FedANAG not only accelerates the global convergence by reducing the required number of communication rounds to a target accuracy, but converges to an eventually higher accuracy.

Preliminary feasibility study of a cognitive stimulation therapy programme for older adults with an intellectual disability.

Despite the increased risk for people with an intellectual disability developing dementia, post-diagnostic psychosocial supports such as cognitive stimulation therapy (CST) are not routinely offered and there is limited research examining this intervention with people with intellectual disabilities. The aim of this study was to explore the feasibility of CST for older adults with intellectual disability to support active ageing. Five client participants attended a 14-session CST group and four staff attended a focus group. Reflexive thematic analysis was used to investigate the client and staff narratives. Three key themes were generated: (1) Brain Health, (2) Connecting with others, and (3) Barriers and Enablers. Findings indicated the suitability of CST as a way of supporting active ageing for older adults with intellectual disability. This study adds to the growing knowledge about service provision for older adults and their changing needs as they age and identifies clinical implications such as staff training to support intervention adherence.

Generalized Federated Learning via Gradient Norm-Aware Minimization and Control Variables

Federated Learning (FL) is a promising distributed machine learning framework that emphasizes privacy protection. However, inconsistencies between local optimization objectives and the global objective, commonly referred to as client drift, primarily arise due to non-independently and identically distributed (Non-IID) data, multiple local training steps, and partial client participation in training. The majority of current research tackling this challenge is mainly based on the empirical risk minimization (ERM) principle, while giving little consideration to the connection between the global loss landscape and generalization capability. This study proposes FedGAM, an innovative FL algorithm that incorporates Gradient Norm-Aware Minimization (GAM) to efficiently search for a local flat landscape. FedGAM specifically modifies the client model training objective to simultaneously minimize the loss value and first-order flatness, thereby seeking flat minima. To directly smooth the global flatness, we propose the more significant FedGAM-CV, which employs control variables to correct local updates, guiding each client to train models in a globally flat direction. Experiments on three datasets (CIFAR-10, MNIST, and FashionMNIST) demonstrate that our proposed algorithms outperform existing FL baselines, effectively finding flat minima and addressing the client drift problem.

Demand study of freight transportation via railway of four commodity groups: A case study on Costa Rica's Pacific Coast Route

Having railway mobility as an option among transportation modes is of prime importance for the freight logistics of a country and the overall performance of its commerce sector. This work presents a case study in Costa Rica, where freight transportation via railway was formerly available during the last century but has not been in full service since 1995. Specifically, this paper assesses the link connecting the capital city of San Jose to the port of Caldera on the Pacific Ocean coast, henceforth referred to as the “Pacific Coast Route”. Although government institutions and industries alike have stressed the importance of its reactivation in recent years, few efforts have been made to specifically study this route with a level of detail that enables them to understand the composition of freight markets.This study presents a novel demand estimation differentiated by four types of commodities developed to calculate the potential amount of goods that could flow through this corridor, measured in monetary terms and as volumes. Trends in the expected share were analyzed to identify variables that may improve the attractiveness of the railway service and the participation of clients along the route from the country's main economic activities which includes the following: agriculture & livestock, manufacturing, minerals, and forestry.The four-step transportation model was chosen for its suitability towards this case study; where available data sets were limited, the information was aggregated into yearly values, and the network was defined with few transportation zones. Activity-based models were not feasible in this case due to the lack of information on complete routes and volume flows. The model evaluated a first scenario given the present design of the route and a second scenario involving improvements in the alignment and the addition of a new segment.Regarding the opportunities of moving goods according to their type, in terms of volume, minerals (48 %) and agricultural (31 %) have the largest shares of participation in this region, followed by manufacturing (20 %) and forestry products (1 %). However, the manufacturing sector stands out in financial terms, accounting for 71 % of the total value of goods mobilized in the study area.It was found that the demand is most sensible for changes in the service fare, while the benefits of shorter travel times and a larger attraction area in scenario 2 were not significant. A service fee of $USD 2.35/ton in average is recommended to reach a 30 % modal shift towards the railway (following the European Commission's goals to reduce CO2 emissions from freight transportation in 2030).Finally, recommendations for public policies are provided to the entities in the country in the sectors of Agriculture, Health, Energy and Environment, Transportation, and Public Services. The suggestions aim to develop the new railway project with a holistic perspective, considering the inclusion of less profitable economic activities of the surrounding communities that require connectivity, compliance with environmental commitments, and design for resilience while aiming for a financially feasible business model.

Testing the incremental effectiveness of pay-for-performance to improve implementation of a motivational interviewing brief intervention for substance use disorders in HIV settings: Results of a parallel cluster-randomized type 3 hybrid trial.

Substance use disorders (SUDs) have a serious adverse impact on the physical and mental well-being of people with HIV. Previously, using a 39-site dual-randomized type 2 hybrid trial design, findings from the Substance Abuse Treatment to HIV Care Project supported the Implementation and Sustainment Facilitation (ISF) strategy to improve implementation and effectiveness of a motivational interviewing brief intervention (MIBI) for SUD within HIV service settings across the United States (US). Building on this trial, this parallel cluster-randomized type 3 hybrid trial aimed to test the incremental effectiveness of a pay-for-performance (P4P), a form of the "alter incentive/allowance structures" strategy. Twenty-six HIV service organizations, their staff participants (N=87), and their client participants (N=341) were cluster-randomized to one of two implementation conditions. The control condition included staff-focused training, feedback, and consultation (TFC) and team-focused implementation and sustainment (ISF). The experimental condition included TFC+ISF as well as P4P (TFC+ISF+P4P). P4P used financial incentives to reward MIBI implementation (US$10 per MIBI delivered) and MIBI implementation at or above a pre-defined level of quality (US$10 per demonstration). In addition to these outcomes, past 4-week changes/reductions in client participant's days of primary substance use and anxiety symptoms were examined. The addition of P4P had a large and significant effect on the number of MIBIs implemented (d=1.30, p<.05) and reduction in anxiety (d=-1.54), but there was no impact on days of substance use. P4P had large effects on MIBI quality (d=1.24) and MIBI implementation effectiveness (d=1.28), but these were not significant (p<.10). P4P is a form of the "alter incentive/allowance structures" strategy. Its function is to reward the implementation of a clinical innovation. Rewarding implementation is consistent with the theory of implementation effectiveness, which suggests implementation climate (i.e., the extent to which implementation is expected, supported, and rewarded) is a key antecedent of implementation effectiveness (i.e., the consistency and quality of implementation). We found that P4P had a significant, positive impact on MIBI implementation in HIV service settings, but client-level outcomes were mixed. Future research should examine the effectiveness of the P4P strategy to improve implementation and sustainment of other evidence-based innovations. ClinicalTrials.gov: NCT04687917. Registered 12/18/2020.

Unmet Caregiver Needs After Traumatic Brain Injury: A Study Exploring OT’s Important Role in Rural Communities

Abstract Date Presented 03/23/24 This study investigated unmet needs of caregivers of individuals with traumatic brain injury (TBI) living in rural communities. Findings provide evidence to support the importance of OT interventions to address caregiver needs and client participation in rural communities. Primary Author and Speaker: Steven Wheeler Additional Authors and Speakers: Amanda Acord-Vira

A Fair Contribution Measurement Method for Federated Learning.

Federated learning is an effective approach for preserving data privacy and security, enabling machine learning to occur in a distributed environment and promoting its development. However, an urgent problem that needs to be addressed is how to encourage active client participation in federated learning. The Shapley value, a classical concept in cooperative game theory, has been utilized for data valuation in machine learning services. Nevertheless, existing numerical evaluation schemes based on the Shapley value are impractical, as they necessitate additional model training, leading to increased communication overhead. Moreover, participants' data may exhibit Non-IID characteristics, posing a significant challenge to evaluating participant contributions. Non-IID data have greatly affected the accuracy of the global model, weakened the marginal effect of the participants, and led to the underestimated contribution measurement results of the participants. Current work often overlooks the impact of heterogeneity on model aggregation. This paper presents a fair federated learning contribution measurement scheme that addresses the need for additional model computations. By introducing a novel aggregation weight, it enhances the accuracy of the contribution measurement. Experiments on the MNIST and Fashion MNIST dataset show that the proposed method can accurately compute the contributions of participants. Compared to existing baseline algorithms, the model accuracy is significantly improved, with a similar time cost.

Enhancing Energy Consumption Prediction in Smart Homes: A Convergence-Aware Federated Transfer Learning Approach

Achieving accurate energy consumption prediction can be challenging, particularly in residential buildings, which experience highly variable consumption behavior due to changes in occupants and the construction of new buildings. This variability, combined with the potential for privacy breaches through conventional data collection methods, underscores the need for novel approaches to energy consumption forecasting. The proposed study suggests a new approach to predict energy consumption, utilizing Federated Learning (FL) to train a global model while ensuring local data privacy and transferring knowledge from information-rich to information-poor buildings. The proposed method learns the transferable knowledge from the source building without any privacy leakage and utilizes it for target buildings. Since the performance of the global model could be negatively affected by some participating nodes with poor performance due to noisy or limited data, we propose a client selection strategy on the server based on the normal distribution for choosing the best possible participants for the global model. Our method enables clients to participate selectively in the aggregation procedure to avoid model divergence due to poor performance. The proposed model is evaluated and conducts in-depth analyses of energy consumption patterns. We validate the performance by comparing its MAE, MSE, and R2 values to those of existing traditional and ensemble models. Our findings indicate that the proposed FL-based model with selective client participation outperforms its counterpart methods regarding predictive accuracy and robustness. The source code is available on Github

The Audacity Health Method: Enhancing Therapeutic Understanding and Client Empowerment

The Audacity Health method utilizes the Feynman Technique, a learning method developed by Nobel Prize-winning physicist Richard Feynman. It emphasizes deep understanding and simplification of complex concepts through teaching. This paper explores the application of the technique within the context of psychotherapy to enhance both therapist and client comprehensions of psychological processes. Implementing a step-by-step approach, therapists can break down and demystify intricate behavioural and emotional patterns, facilitating a more profound grasp of factors that cause and maintain the mental illness the individual is seeking therapy to treat. Through a case study example, this paper demonstrates how the method can be employed to clarify the client’s issues, empowering them to gain insight into their mental health and generate solutions for symptoms. The outcomes suggest that the Audacity Health method fosters more effective communication, reinforces therapeutic alliance, and promotes active client participation in treatment. Ultimately, this integration aims to optimize therapeutic efficacy by merging educational simplification with evidence-based psychotherapeutic interventions to advance mental health.

Implementation and Reach of Health Coaching Using Motivational Interviewing to Reduce Cardiovascular Disease Risk in Uninsured Illinois Women.

Cardiovascular disease (CVD) is the leading cause of mortality for US women; lack of health insurance contributes to poor control of risk factors and increased mortality. Health coaching including motivational interviewing can support primary and secondary CVD prevention, but among uninsured women, improving health outcomes is dependent on successfully reaching priority populations. We evaluated the implementation and reach of health coaching with motivational interviewing among clients in the Illinois WISEWOMAN Program (IWP), a CVD screening and risk-reduction program for uninsured women aged 40 to 64. Following CVD screening, motivational interviewing is offered to all IWP clients via four 30-min one-on-one health coaching sessions to offer personalized guidance on setting and achieving health behavior goals. Our analysis included clients from the eight community-based Illinois agencies that implemented IWP from 2019 to 2023. We assessed client demographic and baseline health characteristics among all IWP clients, those who participated in health coaching by attending at least one session, and those who completed health coaching by attending at least three of four sessions. We also assessed health coaching participation and completion by agency and examined agency-specific associations between client characteristics and health coaching participation and completion. Among IWP enrollees (n =3094), 89.7% participated in at least one health coaching session but only 31.4% completed health coaching by attending at least three sessions. Over 90% of IWP clients participated in at least one health coaching session at 4 IWP agencies. Further, over 85% of health coaching participants completed health coaching at four agencies. Across agencies, no client-level characteristics were consistently associated with health coaching participation or completion. High motivational interviewing participation rates support its acceptability among uninsured women, but agency-level community-level barriers likely prevent client engagement in multiple sessions. Reducing CVD risk requires working with partner agencies to address barriers to reaching the priority population.

Giving Voice: Examining the Tactical Repertoires of Nonprofit Advocacy for Disadvantaged Populations

Abstract Representation is an important role of the nonprofit sector as these organizations can give voice to the priorities of their constituents. This is particularly important for disadvantaged communities that may lack access to traditional democratic processes. What advocacy tactics are most utilized by nonprofit organizations serving disadvantaged populations? Given variation in nonprofit advocacy tactics, which tactics are perceived as most effective? To answer these questions, we look to the framework of tactical repertoires from sociological studies of social movements. Analysis of a survey of New England nonprofits (N = 656, 55 % response rate) indicates that nonprofits serving disadvantaged populations use similar tactics to organizations advocating for other constituencies but employ them at higher rates. These tactics cluster into three tactical repertoires, each a different combination of advocacy tactics. Perceptions of advocacy effectiveness are associated with the size and nature of tactical repertoires, client participation in advocacy, government funding, and perceptions of the policy environment.

Eco-FL: Enhancing Federated Learning sustainability in edge computing through energy-efficient client selection

In the realm of edge cloud computing (ECC), Federated Learning (FL) revolutionizes the decentralization of machine learning (ML) models by enabling their training across multiple devices. In this way, FL preserves privacy and minimizes the need for centralized data by processing data near the source. From a communication standpoint, only the model weights are exchanged between devices. By avoiding the need to send data to a centralized location for processing, FL reduces the energy required for data transfer and supports more efficient use of computing resources at the edge. FL is particularly advantageous for resource-constrained devices, such as smartphones and IoT devices. However, this limited computational power and battery capacity and the challenge of energy consumption are critical aspects of FL systems. This paper introduces Eco-FL, an innovative methodology designed to optimize energy consumption in FL systems, in the field of Green Edge Cloud Computing (GECC). Our approach employs a device selection process that considers the entropy of the data held by the devices and their available energy reserves. This ensures that devices with lower energy availability are less likely to participate in the training rounds, prioritizing those with higher energy capacities. To evaluate the efficacy of our methodology, we utilize FedEntropy, an entropy-based aggregation method, alongside established aggregation methods such as FedAvg and FedProx for performance comparison. The effectiveness of Eco-FL in reducing energy consumption without compromising the accuracy of the FL process is demonstrated through analyses conducted on three distinct datasets. These analyses vary the β parameter of the Dirichlet distribution and account for scenarios with both homogeneous and heterogeneous initial device charges. Our findings validate Eco-FL’s potential to enhance the sustainability of FL systems by judiciously managing client participation based on energy criteria, presenting a significant step forward in the development of energy-efficient FL.

Self-balancing Incremental Broad Learning System with privacy protection

Incremental learning algorithms have been developed as an efficient solution for fast remodeling in Broad Learning Systems (BLS) without a retraining process. Even though the structure and performance of broad learning are gradually showing superiority, private data leakage in broad learning systems is still a problem that needs to be solved. Recently, Multiparty Secure Broad Learning System (MSBLS) is proposed to allow two clients to participate training. However, privacy-preserving broad learning across multiple clients has received limited attention. In this paper, we propose a Self-Balancing Incremental Broad Learning System (SIBLS) with privacy protection by considering the effect of different data sample sizes from clients, which allows multiple clients to be involved in the incremental learning. Specifically, we design a client selection strategy to select two clients in each round by reducing the gap in the number of data samples in the incremental updating process. To ensure the security under the participation of multiple clients, we introduce a mediator in the data encryption and feature mapping process. Three classical datasets are used to validate the effectiveness of our proposed SIBLS, including MNIST, Fashion and NORB datasets. Experimental results show that our proposed SIBLS can have comparable performance with MSBLS while achieving better performance than federated learning in terms of accuracy and running time.

Beyond Open Innovation: scientific mapping and the synthesis of the Co-innovation process as a new paradigm

Purpose: Mapping the theoretical framework of publications on Open Innovation (OI) and Co-innovation regarding the concept of value and synthesizing the literature on Co-innovation, seeking the flow of contributions and proposing a research agenda. Design/Methodology/Approach: A bibliometric analysis and integrative review was carried out through articles, reviews and conference articles published in international databases according to Prado et al. (2016) and Torraco (2016). Originality: Studies have revisited, refined and criticized OI over the years, whose theoretical field has advanced and debated the emergence of new concepts, such as Co-innovation. However, the discussion about how this new paradigm emerges and dialogues with OI and can be considered an evolutionary theme is still lacking in the literature. Results: Although publications and temporal trends on Co-innovation have shown to be promising, the highlighted authors did not present temporal extension of their productions and the field is still in constant oscillations. Despite the streams of contributions with important scope in the construction of the paradigm, the theme still needs epistemological discussions for theoretical consolidation. Theoretical and methodological contributions: This study contributes to epistemological advances on OI and stimulates reflection for discussions and development of contemporary approaches to innovation. Management contributions: This research presents dimensions and stages of Co-innovation processes with elements that impact its practice. Furthermore, it sheds light on the importance of selecting partners (identifying different actors and their roles in the co-creation process) and establishing adequate mechanisms to encourage client participation.

FedUB: Federated Learning Algorithm Based on Update Bias

Federated learning, as a distributed machine learning framework, aims to protect data privacy while addressing the issue of data silos by collaboratively training models across multiple clients. However, a significant challenge to federated learning arises from the non-independent and identically distributed (non-iid) nature of data across different clients. non-iid data can lead to inconsistencies between the minimal loss experienced by individual clients and the global loss observed after the central server aggregates the local models, affecting the model’s convergence speed and generalization capability. To address this challenge, we propose a novel federated learning algorithm based on update bias (FedUB). Unlike traditional federated learning approaches such as FedAvg and FedProx, which independently update model parameters on each client before direct aggregation to form a global model, the FedUB algorithm incorporates an update bias in the loss function of local models—specifically, the difference between each round’s local model updates and the global model updates. This design aims to reduce discrepancies between local and global updates, thus aligning the parameters of locally updated models more closely with those of the globally aggregated model, thereby mitigating the fundamental conflict between local and global optima. Additionally, during the aggregation phase at the server side, we introduce a metric called the bias metric, which assesses the similarity between each client’s local model and the global model. This metric adaptively sets the weight of each client during aggregation after each training round to achieve a better global model. Extensive experiments conducted on multiple datasets have confirmed the effectiveness of the FedUB algorithm. The results indicate that FedUB generally outperforms methods such as FedDC, FedDyn, and Scaffold, especially in scenarios involving partial client participation and non-iid data distributions. It demonstrates superior performance and faster convergence in tasks such as image classification.