Decentralised facility-based training as an alternative model for SLMTA implementation: The Cameroon experience.

IntroductionThe Amana Regional Hospital Laboratory in Tanzania was selected, along with 11 other regional and district laboratories, to participate in a pilot programme for laboratory quality improvement using the Strengthening Laboratory Management Toward Accreditation (SLMTA) training programme.Programme implementationThe SLMTA programme entailed hands-on learning, improvement projects between and after a three-workshop series, supervisory visits from an oversight team and an expert laboratory mentor to facilitate and coach the process. Audits were conducted at baseline, exit (approximately one year after baseline) and follow-up (seven months after exit) using the Stepwise Laboratory Quality Improvement Process Towards Accreditation (SLIPTA) checklist. Quality stars (zero to five) were awarded based on audit scores.ResultsWith a dedicated staff and strong leadership from laboratory management, Amana Laboratory implemented processes, policies and procedures recommended as elements of best laboratory practices. The laboratory improved from zero stars (36%) at baseline to successfully achieving three stars (81%) at exit. This was the highest score achieved by the 12 laboratories in the programme (the median exit score amongst the other laboratories was 58%). Seven months after completion of the programme, the laboratory regressed to one star (62%).DiscussionAs the SLMTA improvement programme progressed, Amana Laboratory’s positive attitude and hard work prevailed. With the assistance of a mentor and the support of the facility’s management a strong foundation of good practices was established. Although not all improvements were maintained after the conclusion of the programme and the laboratory dropped to a one-star rating, the laboratory remained at a higher level than most laboratories in the programme.

BackgroundSince its introduction in 2009, the Strengthening Laboratory Management Toward Accreditation (SLMTA) programme has been implemented in 617 laboratories in 47 countries.ObjectiveWe completed a systematic review of the published literature on SLMTA. The review consists of two companion papers; this article examines quantitative evidence presented in the publications along with a meta-analysis of selected results.MethodsWe identified 28 published articles with data from SLMTA implementation. The SLMTA programme was evaluated through audits based on a standard checklist, which is divided into 12 sections corresponding to the 12 Quality System Essentials (QSEs). Several basic service delivery indicators reported by programmes were also examined. Results for various components of the programme were reviewed and summarised; a meta-analysis of QSE results grouped by the three stages of the quality cycle was conducted for 126 laboratories in 12 countries.ResultsGlobal programme data show improved quality in SLMTA laboratories in every country, with average improvements on audit scores of 25 percentage points. Meta-analysis identified Improvement Management as the weakest stage, with internal audit (8%) and occurrence management (16%) showing the lowest scores. Studies documented 19% – 95% reductions in turn-around times, 69% – 93% reductions in specimen rejection rates, 76% – 81% increases in clinician satisfaction rates, 67% – 85% improvements in external quality assessment results, 50% – 66% decreases in nonconformities and 67% increases in staff punctuality.ConclusionsThe wide array of results reported provides a comprehensive picture of the SLMTA programme overall, suggesting a substantive impact on provision of quality laboratory services and patient care. These comprehensive results establish a solid data-driven foundation for program improvement and further expansion.

The Strengthening Laboratory Management Toward Accreditation (SLMTA) program was developed to promote immediate, measurable improvement in laboratories of developing countries. The laboratory management framework, a tool that prescribes managerial job tasks, forms the basis of the hands-on, activity-based curriculum. SLMTA is implemented through multiple workshops with intervening site visits to support improvement projects. To evaluate the effectiveness of SLMTA, the laboratory accreditation checklist was developed and subsequently adopted by the World Health Organization Regional Office for Africa (WHO AFRO). The SLMTA program and the implementation model were validated through a pilot in Uganda. SLMTA yielded observable, measurable results in the laboratories and improved patient flow and turnaround time in a laboratory simulation. The laboratory staff members were empowered to improve their own laboratories by using existing resources, communicate with clinicians and hospital administrators, and advocate for system strengthening. The SLMTA program supports laboratories by improving management and building preparedness for accreditation.

BackgroundIn 2010, a National Laboratory Strategic Plan was set forth in Ethiopia to strengthen laboratory quality systems and set the stage for laboratory accreditation. As a result, the Strengthening Laboratory Management Toward Accreditation (SLMTA) programme was initiated in 45 Ethiopian laboratories.ObjectivesThis article discusses the implementation of the programme, the findings from the evaluation process and key challenges.MethodsThe 45 laboratories were divided into two consecutive cohorts and staff from each laboratory participated in SLMTA training and improvement projects. The average amount of supportive supervision conducted in the laboratories was 68 hours for cohort I and two hours for cohort II. Baseline and exit audits were conducted in 44 of the laboratories and percent compliance was determined using a checklist with scores divided into zero- to five-star rating levels.ResultsImprovements, ranging from < 1 to 51 percentage points, were noted in 42 laboratories, whilst decreases were recorded in two. The average scores at the baseline and exit audits were 40% and 58% for cohort I (p < 0.01); and 42% and 53% for cohort II (p < 0.01), respectively. The p-value for difference between cohorts was 0.07. At the exit audit, 61% of the first and 48% of the second cohort laboratories achieved an increase in star rating. Poor awareness, lack of harmonisation with other facility activities and the absence of a quality manual were challenges identified.ConclusionImprovements resulting from SLMTA implementation are encouraging. Continuous advocacy at all levels of the health system is needed to ensure involvement of stakeholders and integration with other improvement initiatives and routine activities.

BackgroundIn 2011 the Stepwise Laboratory Quality Improvement Process Towards Accreditation (SLIPTA) was launched, aimed at strengthening the quality and competence of African clinical, public health and reference laboratories. We reviewed the first version of the SLIPTA checklist in 2011. The continued development and publication of a new version of the International Organization for Standardization (ISO) 15189 standard demands a renewed review.ObjectiveThis study aimed to determine the suitability of SLIPTA in guiding laboratories towards ISO 15189:2012 compliance and accreditation and provide recommendations for further SLIPTA improvement.MethodsThe study was conducted between September 2018 and April 2019. Coverage of ISO 15189:2012 by SLIPTA checklist version 2:2015 was determined and the point distribution of the scoring system over the different sections of the SLIPTA checklist was re-investigated. These findings were compared with the review of the first version of the SLIPTA checklist (based on ISO 15189:2007) and with findings published on SLIPTA implementation and roll-out.ResultsThe coverage of ISO 15189 by the SLIPTA checklist has increased, even though ISO 15189:012 is more extensive than ISO 15189:2007. The point distribution is still skewed towards sections related to quality planning rather than quality control and improvement. Although to date 314 laboratories have been assessed, barriers for laboratories to participate in SLIPTA are high. Sustainability of SLIPTA results is insufficiently studied.ConclusionSLIPTA checklist version 2:2015 has improved compared to earlier versions. We recommend increasing accessibility for laboratories to participate and increasing guidance for ISO-based quality management system implementation.

BackgroundLaboratory Quality Management Systems (LQMS), antimicrobial resistance (AMR) surveillance and stewardship (AMS) are important for quality patients’ care and safety. Through collaboration between Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) and...

The orientation of music teachers’ talent training and the construction of talent training models are the core issues of concern for the reform of music education majors. Facing the new requirements of society for music education professional talents training in the new period, this paper proposes an improved talent training model based on the ADDIE model by studying the current situation of music education professional talents training from the perspective of adapting to the development needs of music curriculum reform in basic education. From the assessment of students’ and teachers’ evaluation and acceptance of the model, the mean value of students’ evaluation scores increased from 64.7 to 75.7 compared with the original training model, an increase of 17.00%. The mean teacher evaluation score increased from 67.0 to 72.7, an increase of 8.51%. It is evident that the talent development model has improved and is accepted by students and teachers. In terms of the acceptance of the training model proposed in this paper, the percentage of students who strongly agree with it has increased by 19.8%, the percentage of students who can accept it has increased by 20.2%, and the percentage of students who cannot adapt to the teaching model has decreased by 6.1 percentage points, with a comparative decrease of 20.3%. Teachers were also more accepting of the new training model, and the percentage of those who could accept it increased by 5.3 percentage points and 7.17% in comparison. In conclusion, the talent cultivation model of music education based on the ADDIE model can be well adapted to the teaching classroom of colleges and universities and also meets the needs of the country and society for music teachers in the new era, which is conducive to the cultivation of composite talents who are highly educated and have a solid theoretical foundation and excellent practical ability.

Extension services are a keystone of information diffusion in agriculture. This paper exploits a large randomized controlled trial to track diffusion of a new technique in the classic Training and Visit (T&V) extension model, relative to a more direct training model. In both control and treatment communities, contact farmers (CFs) serve as points-of-contacts between agents and other farmers. The intervention (Treatment) aims to address two pitfalls of the T&V model: i) infrequent extension agent visits, and ii) poor quality information. Treatment CFs receive a direct, centralized training. Control communities are exposed to the classic T&V model. Information diffusion was tracked through two nodes: from agents to CFs, and from CFs to others. Directly training CFs leads to large gains in information diffusion and adoption, and CFs learn by doing. Diffusion to others is limited: other males adopt the technique perceived as labor saving, with an effect size of 75 percent.

This study presents a privacy-preserving learning model designed for cross-border telemedicine in East Africa that keeps raw patient records in country while hospitals collaborate on model quality. The core of this approach is to keep sensitive patient records localized within each country, with hospitals training models locally and only sharing model updates. Using synthetic EHRs split across seven hospitals in Kenya, Tanzania, and Uganda, we compare centralized training, standard federated learning, and federated learning with differential privacy. Federated learning improves utility while maintaining data localization, with accuracy rising by about 0.0665, recall for the positive class improving by about 0.1193, and F1 increasing by about 0.0657 relative to centralized training. Adding differential privacy made the system more resilient to attacks. The success rate of model-inversion attacks dropped from 0.696 in the centralized training scenario to 0.686 with standard FL and further to 0.638 with FL + DP. This represents an absolute reduction of 0.058, or about 8.4 percent, in attack success. Membership-inference leakage has an AUC of around 0.50. The trade-off is tunable utility at a chosen privacy budget, for example accuracy near 0.530 at ε = 0.30. The originality is practical, we pair federated learning with an attack simulator and an ε register that turns privacy into an auditable setting hospitals can manage during cross-border care.

Tailgating, which occurs when a vehicle follows another at an unsafe distance, is a major contributor to rear-end collisions, especially on roads with dense traffic, frequent lane changes, and limited infrastructure, creating hazardous conditions. Modern intelligent transportation systems and connected vehicle technologies help monitor and reduce unsafe behaviours. Still, conventional deep learning methods rely on centralised training, which raises concerns about privacy, scalability, and delays. This framework integrates collaborative training with federated learning and a bidirectional long short-term memory network to detect tailgating accurately. Realistic vehicle trajectory data were generated using the Simulation of Urban mobility tool to replicate diverse driving scenarios, including urban and highway conditions. Key features such as following distance, relative speed, braking distance, time headway, vehicle speed, and driver reaction distance were derived to model driving behaviour. Baseline models, including convolutional neural networks, recurrent neural networks, and long short-term memory networks, were developed for comparison. The results show that the framework combining federated learning and a bidirectional long short-term memory network achieved detection accuracy above 91.98%, outperforming centralised training and baseline models. The study shows that a collaborative, decentralised approach can capture bidirectional temporal dependencies in driving sequences, ensuring driving data remains local while supporting real-time detection.

Blended digital mental health interventions combining technology with human support are more effective than stand-alone treatments. However, limited research has examined how to train and supervise personnel delivering human support components. The Kuamsha app, a gamified digital intervention for adolescent depression based on behavioral activation, was designed to be paired with low-intensity telephone-based peer support. A structured training and supervision program for peer supporters was codeveloped through workshops with mental health professionals and youth with lived experience of mental health challenges in South Africa and Uganda. To the best of our knowledge, this is the first study to evaluate a structured peer mentor model within a digital mental health intervention in low- and middle-income countries. This study assessed the feasibility, acceptability, and fidelity of a training and supervision program for peer supporters delivering a digital mental health intervention in South Africa and Uganda. We conducted a mixed methods evaluation of the peer mentor program. Quantitative metrics assessed the feasibility of recruitment, retention, and attendance among peer mentors (n=13, South Africa; n=4, Uganda), as well as training acceptability. Fidelity, adherence, and competence were scored at the session level and converted to percentages of the maximum possible score. Linear mixed-effects regression models with a random intercept for provider and site estimated adjusted marginal means (95% CI). In-depth interviews and focus group discussions explored program acceptability and implementation factors. The peer mentor training and supervision program was feasible and acceptable in both settings, with high recruitment (South Africa: n=13/19, 68%; Uganda: 4/4, 100%), retention (South Africa: 9/13, 69%; Uganda: 4/4, 100%), and training attendance rates (89%-92% in South Africa and 100% in Uganda), alongside qualitative reports of high satisfaction. All peer mentors met a minimum posttraining competency threshold (≥50%), with median competency scores of 70.7% (IQR 45.8%-78.2%) in South Africa and 75.4% (IQR 73.8%-77.3%) in Uganda. Independent ratings of recorded calls indicated high overall fidelity in South Africa (84.7%, 95% CI 80.3%-89.0%) and Uganda (87.7%, 95% CI 83.4%-92.1%). Adherence was higher in Uganda than South Africa (adjusted mean difference [AMD] 13.30 percentage points, 95% CI 8.99-17.61; P<.001), as was competence (AMD 4.88 percentage points, 95% CI 1.23-8.53; P=.009). The AMD in overall fidelity (3.06 percentage points, 95% CI -0.98 to 7.10) was not statistically significant (P=.14). The qualitative findings emphasized the value of ongoing supervision and capacity development, interactive training approaches, and blended delivery models. Locally adapted training and supervision models can strengthen peer mentor capabilities to support digital interventions. Adequate supervisory capacity and incentive structures are critical to sustain engagement, retention, and fidelity. In settings with frequent network disruptions, periodic in-person contact between peer mentors and supervisors may enhance fidelity. Future research should examine how peer mentor fidelity influences user engagement and mental health outcomes.

Additive manufacturing (AM) enables the production of customized and complex metal parts, providing greater flexibility than traditional mass production. However, maintaining consistent product quality for AM processes remains challenging due to process variability and heat emission fluctuations, which can cause defects such as overheating. Overheating occurs when non-uniform temperature distributes across the printed parts, leading to issues such as structural weaknesses. Predicting heat emissions accurately is essential to preventing such issues, but AM facilities, particularly small- and medium-sized enterprises (SMEs), generate limited and heterogeneous datasets due to unique configurations and privacy concerns, making traditional centralized predictive model training paradigms impractical. Federated learning (FL) offers a privacy-preserving solution, it alleviates data privacy and data scarcity concerns by enabling collaborative learning across decentralized datasets while keeping data local. However, its performance suffers under heterogeneous client data and varied computational resources. This creates a need to address data heterogeneity while maintaining both global model performance and individual client personalization. We propose FedScope-KD, a novel heterogeneity-aware FL framework designed to balance the scope of global and personalized learning. Client heterogeneity level is quantified prior to the FL training, to improve the model weight aggregation during training. This is achieved through component decomposition and knowledge distillation. The framework decomposes client data into process-invariant representations, which capture the fundamental nature of the laser powder bed fusion (LPBF) process, and process-variant representations, which reflect local, client-specific configuration settings. This decomposition ensures that global models benefit from collaborative learning on the homogeneous data representations, while personalized predictions remain tailored to client-specific heterogeneous data representations. A knowledge distillation (KD) mechanism further enhances the framework by efficiently transferring process-invariant knowledge from a global teacher model to local student models. We compare the heterogeneity index generation and calculate the Dynamic Time Warping (DTW) distance for both inter-client variant and invariant components. The variant components exhibit greater distances between clients, reflecting their inherent heterogeneity, while the invariant components are more closely aligned across clients. This pattern demonstrates the effectiveness of the decomposition process in separating heterogeneous and homogeneous information. The heterogeneity index is further used for weight aggregation during FL training. The framework’s performance is validated through a comparison with baseline FL methods such as FedAvg, as well as centralized and individual learning paradigms. Performance metrics include personalized performance and global accuracy with MSE. Through experiments on heat emission datasets in the LPBF process, we show that FedScope-KD successfully addresses the challenges of data heterogeneity and resource constraints, providing a robust solution of predictive modeling for SMEs.

Bias in medical image segmentation can lead to unequal performance across demographic subgroups, raising concerns about fairness and reliability in clinical AI systems. While deep learning models have achieved high segmentation accuracy, ensuring equitable performance across race and gender remains a significant challenge, particularly in privacy-sensitive healthcare environments. This study investigates fairness-aware medical image segmentation for hip and knee radiographs using deep learning models evaluated in both centralized and Federated Learning (FL) settings. We introduce Curriculum Learning (CL) strategies and Progressive Loss (PL) functions to regulate sample difficulty during training. In addition, we propose two novel fairness-oriented federated learning algorithms, Federated Intersection over Union (FedIoU) and Federated Intersection over Union with Outlier Analysis (FedIoUoutlier). Experiments are conducted using multiple segmentation backbones and simulated multi-site data partitions derived from the Osteoarthritis Initiative dataset. Model performance is evaluated using Intersection over Union (IoU), IoU standard deviation, Skewed Error Ratio (SER), and Min-Max Disparity across race and gender subgroups. Statistical significance was verified using paired t-tests to compare per-sample IoU performance against baseline configurations. Across both hip and knee segmentation tasks, curriculum learning and progressive loss strategies consistently improved segmentation accuracy and reduced demographic performance disparities in centralized training. In federated settings, fairness-aware aggregation further enhanced performance. Notably, FedIoUoutlier combined with balanced curriculum learning and tiered progressive loss achieved the highest mean IoU while yielding the lowest SER and Min-Max Disparity, indicating improved fairness without sacrificing accuracy. In several configurations, federated models matched or exceeded the performance of optimized centralized models, with statistically significant improvements in per-sample IoU over baseline configurations. The results demonstrate that structured training strategies and fairness-aware federated aggregation can jointly improve accuracy, stability, and demographic fairness in medical image segmentation. By integrating curriculum learning, progressive loss, and novel FL algorithms, this work provides a practical pathway toward equitable and privacy-preserving AI systems for medical imaging.

The rapid expansion of Internet of Things (IoT) devices has heightened the need for effective intrusion detection systems (IDS) that operate under strict resource constraints. Conventional IDS implementations require substantial computational resources, making them unsuitable for low-power microcontroller-based devices. This paper proposes a novel collaborative IDS architecture that separates centralised model training from distributed edge inference. The system employs an autoencoder-based labelling mechanism trained on regular traffic to identify anomalies. Each ESP32 device performs local inference and exchanges predictions via UDP multicast, whilst MD5 hashing ensures model consistency across the network. Collaborative verification enables devices to identify and isolate compromised nodes without central coordination. Experimental evaluation demonstrates 98.5% F-score with 3ms average inference latency and 12.7KB memory footprint, consuming only 2.4% of available SRAM. Our approach achieves superior detection accuracy compared to existing cloud-edge systems whilst operating on severely resource-constrained hardware, making it a practical solution for large-scale IoT security deployments.

Insulators are widely used in distribution network transmission lines and serve as critical components of the distribution network. The detection of insulator faults is essential to ensure the safe and stable operation of the distribution network. Traditional insulator detection methods often rely on manual identification, which is time-consuming, labor-intensive, and inaccurate. The use of vision sensors for object detection is an efficient and accurate detection method that requires minimal human intervention. Currently, there is a considerable amount of research on the application of vision sensors for insulator fault recognition in object detection. However, centralized object detection requires uploading data collected from various substations through vision sensors to a computing center, which may raise data privacy concerns and increase uncertainty and operational risks in the distribution network. Therefore, this paper proposes a privacy-preserving insulator detection method based on federated learning. An insulator fault detection dataset is constructed, and Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) models are trained within the federated learning framework for insulator fault detection. Most of the existing insulator anomaly detection methods use a centralized model training method, which has the advantage of achieving a target detection accuracy of over 90%, but the disadvantage is that the training process is prone to privacy leakage and lacks privacy protection capability. Compared with the existing insulator target detection methods, the proposed method can also achieve an insulator anomaly detection accuracy of more than 90% and provide effective privacy protection. Through experiments, we demonstrate the applicability of the federated learning framework for insulator fault detection and its ability to protect data privacy while ensuring test accuracy.