Training Approach Research Articles

The Effect of Electrical Stimulation Strength Training on Lower Limb Muscle Activation Characteristics During the Jump Smash Performance in Badminton Based on the EMS and EMG Sensors.

This study investigates the effects of electrical stimulation (EMS) combined with strength training on lower limb muscle activation and badminton jump performance, specifically during the "jump smash" movement. A total of 25 male badminton players, with a minimum of three years of professional training experience and no history of lower limb injuries, participated in the study. Participants underwent three distinct conditions: baseline testing, strength training, and EMS combined with strength training. Each participant performed specific jump tests, including the jump smash and static squat jump, under each condition. Muscle activation was measured using electromyography (EMG) sensors to assess changes in the activation of key lower limb muscles. The EMS intervention involved targeted electrical pulses designed to stimulate both superficial and deep muscle fibers, aiming to enhance explosive strength and coordination in the lower limbs. The results revealed that the EMS + strength condition significantly improved performance in both the jump smash and static squat jump, as compared to the baseline and strength-only conditions (F = 3.39, p = 0.042; F = 3.67, p = 0.033, respectively). Additionally, increased activation of the rectus femoris (RF) was observed in the EMS + strength condition, indicating improved muscle recruitment and synchronization, likely due to the activation of fast-twitch fibers. No significant differences were found in the eccentric-concentric squat jump (F = 0.59, p = 0.561). The findings suggest that EMS, when combined with strength training, is an effective method for enhancing lower limb explosiveness and muscle activation in badminton players, offering a promising training approach for improving performance in high-intensity, explosive movements.

A distributed training framework for abnormal sound detection in industrial devices: aggregation strategies and performance analysis

Abstract In today's digital and networked industrial landscape, the detection of abnormal sounds in device has emerged as a vital aspect for guaranteeing the normal operation of industrial machinery. Nevertheless, traditional centralized training approaches demand substantial amounts of audio data, imposing considerable burdens on data storage and transmission, and concurrently presenting obstacles to data privacy and security. Federated learning, as a distributed machine learning paradigm, enables model training with local data from each client without sharing the original data, thereby effectively safeguarding data privacy. Hence, in this study, we propose a distributed training framework based on federated training, which enables multiple clients to collaboratively train an abnormal sound detection model, thereby mitigating the risk of data privacy exposure. In the distributed training framework, each client possesses data from different types of device or various machines within the same device type, posing significant challenges for distributed training. To overcome this, we devised two client device distribution scenarios and proposed aggregation strategies based on client sample size, model performance, and domain shift among clients. Additionally, we introduced a Sample-Performance-Shift (SPS) aggregation strategy to ensure robust model performance across diverse device scenarios in industrial settings.
The proposed methods were evaluated on the DCASE 2020 Challenge Task 2 dataset.
Experimental results demonstrate that the SPS aggregation strategy enhances the accuracy and reliability of abnormal sound detection for industrial device within the distributed training framework while simultaneously reducing the risk of data privacy leakage.

Comparison of the effect of quick response code-based training and traditional methods on the knowledge, skill, and satisfaction of the nursing students in using medical device

Nurses acknowledge that their understanding of the shortcomings in the utilization of medical devices stems from insufficient knowledge about their correct usage. The use of Quick Response (QR) technology has paved a new gateway for accessing information and resources. The available evidence confirmed the usefulness of QR codes for use in clinical settings as a tool for training. The present research performed a quasi-experimental research design to compare the effect of QR code-based training and traditional methods on the knowledge, skill, and satisfaction of the nursing students in using medical device. For the control group, textual Persian manuals, visual aids in the form of pictures, and URLs were available to the students. For the intervention group, the training approach based on QR codes was explained on the first day of their presence at the heart center. The knowledge, skill and satisfaction evaluations were conducted at the end of the clinical course on day 10. The group of nursing students who used the QR code-based training method had a statistically significant higher mean knowledge and skill score compared to the traditional training method group. The level of satisfaction among students who received training using QR codes was non-significantly higher than traditional method group. Nursing managers and educators can enhance the delivery of high-quality nursing care by using QR code.

Evaluation of Different Few-Shot Learning Methods in the Plant Disease Classification Domain.

Early detection of plant diseases is crucial for agro-holdings, farmers, and smallholders. Various neural network architectures and training methods have been employed to identify optimal solutions for plant disease classification. However, research applying one-shot or few-shot learning approaches, based on similarity determination, to the plantdisease classification domain remains limited. This study evaluates different loss functions used in similarity learning, including Contrastive, Triplet, Quadruplet, SphereFace, CosFace, and ArcFace, alongside various backbone networks, such as MobileNet, EfficientNet, ConvNeXt, and ResNeXt. Custom datasets of real-life images, comprising over 4000 samples across 68 classes of plant diseases, pests, and their effects, were utilized. The experiments evaluate standard transfer learning approaches alongside similarity learning methods based on two classes of loss function. Results demonstrate the superiority of cosine-based methods over Siamese networks in embedding extraction for disease classification. Effective approaches for model organization and training are determined. Additionally, the impact of data normalization is tested, and the generalization ability of the models is assessed using a special dataset consisting of 400 images of difficult-to-identify plant disease cases.

Multi-Target Detection Method of Intelligent Driving Traffic Scene Based on Faster R-CNN++

The current multi-objective detection methods are plagued by several issues, including insufficient detection efficiency, slow speeds, and difficulties in deciphering driving rules. To address these challenges, we introduce a traffic scene classification method that utilizes Improved Faster Regions with Convolutional Neural Network features (Faster R-CNN++). Initially, the shared convolutional layer and the Recurrent Criss-Cross Attention (RCCA) layer are employed to extract features from the input image. Subsequently, the derived feature map is fed into the Region Proposal Network (RPN) to generate detection boxes, and the Region of Interest Align (RoI Align) layer selects features corresponding to each Region of Interest (RoI) on the feature map, guided by the RPN’s output. Ultimately, we adopt an alternating training approach. Simulation experiments confirm the effectiveness of our method in multi-scene object detection. Our method has demonstrated significant improvements over existing algorithms and has delivered outstanding performance on the BDD-100[Formula: see text]K, ApolloScape, and NuScenes datasets. The results indicate that our deep learning-based traffic scene classification method can accurately discern the behavioral characteristics of various traffic participants.

Solution-Focused Telemental Health: A New Approach for Social Work Field Education

ABSTRACT The COVID-19 pandemic has brought unprecedented challenges around the world and shifted social work education. The solution-focused telemental health (SFTH) model was developed and evaluated at a social work program in New York State in the United States with both Master and Bachelor of Social Work students. The 14-week online training model included: (a) a clinical weekly seminar augmenting social work students’ practice classes focusing on solution-focused brief therapy (SFBT), (b) group supervision and case consultations, and (c) a field placement implementing SFTH. A cotherapy model was implemented where students worked in pairs for additional support, processing, and preparation. Process evaluation using focus groups with social work students, and their process recordings and weekly supervision logs revealed: (a) increased social work students’ competence in using SFBT in their telemental health practice, (b) strong collaboration with peers in the virtual training with group cohesion and mutual support, and (c) needs for more focused and meaningful telemental health training. Recommendations are provided for social work education to incorporate training in telemental health and virtual social work practice using evidence-based approaches. The SFTH model provides an effective, short-term, online training approach for social work students.

Factors Associated with Nonspecialist Quality of Delivery within a Family Strengthening Intervention in Rwanda: a Parallel Latent Growth Model.

In task-shared, mental health, and psychosocial support interventions, monitoring the quality of delivery (fidelity and competence) of nonspecialist providers is critical. Quality of delivery is frequently reported in brief, summary statistics, and while both fidelity and competence scores tend to be high, rarely have factors associated with quality of delivery in low-resource, mental health, and psychosocial support interventions been examined using inferential statistics. Understanding both modifiable and non-modifiable predictors of quality of delivery is important for adapting training and supervision approaches throughout intervention delivery. In this study, we use a parallel process latent growth model to examine the association of non-modifiable, demographic characteristics of nonspecialists and changes in both fidelity and competence over time. We find that nonspecialist age is significantly associated with higher initial fidelity and competence scores and smaller improvements in fidelity and competence over time, although this finding is interpreted in the presence of ceiling effects. In addition, nonspecialists in a certain district were more likely to have higher initial fidelity and competence scores but also see smaller changes over time. Fidelity and competence were found to significantly co-vary. This study provides conceptual and measurement guidance regarding quality of delivery, suggesting that fidelity and competence are theoretically distinct and must be measured separately, but linked together under the umbrella of quality of delivery. This study also has implications for recruiting, training, and supporting nonspecialists delivering behavioral interventions, suggesting that future implementation teams can further contribute to research on how to better support high-quality training, supervision, and personal and professional growth among the growing nonspecialist workforce globally.

Math misconceptions: Mistakes, misunderstanding, and confusion

In this study, we investigate mathematical misconceptions, particularly prevalent among high school students, and offer solutions. The primary aims of our research include identifying the most common mathematical misconceptions, uncovering their underlying causes, and evaluating research-based solutions that have successfully addressed these issues. For this study, data were collected using a mixed-methods approach in Stirling Schools operating throughout Iraq. Data were gathered via a diagnostic test, as well as teacher interviews and classroom observations. The test revealed conceptual, procedural, and application problems, while interviews and observations gave qualitative information about the sources of misunderstandings and teaching tactics. The data analysis blended quantitative mistake classification with qualitative theme analysis to provide a complete picture of misunderstandings and effective training approaches. In light of the data obtained, we can state that misconceptions typically stem from a lack of conceptual understanding, over-reliance on procedural methods, and contextual misunderstandings. Additionally, it has been shown that conceptual education, visual aids, and peer-supported learning help eliminate mathematical misconceptions. These valuable findings hold significant implications for educators seeking to improve students' mathematical knowledge by emphasizing the importance of concept-based learning and providing opportunities for re-education. Through this study, we offer key insights into how widespread mathematical misconceptions can be effectively addressed, contributing meaningfully to mathematics education.

You Don’t Have to Say Where to Edit! jLED – Joint Learning to Localize and Edit Source Code

Learning to edit code automatically is becoming more and more feasible. Thanks to recent advances in Neural Machine Translation (NMT), various case studies are being investigated where patches are automatically produced and assessed either automatically (using test suites) or by developers themselves. An appealing setting remains when the developer must provide a natural language input of the requirement for the code change. A recent proof of concept in the literature showed that it is indeed feasible to translate these natural language requirements into code changes. A recent advancement, MODIT [8], has shown promising results in code editing by leveraging natural language, code context, and location information as input. However, it struggles when location information is unavailable. While several studies [29, 81] have demonstrated the ability to edit source code without explicitly specifying the edit location, they still tend to generate edits with less accuracy at the line level. In this work, we address the challenge of generating code edits without precise location information, a scenario we consider crucial for the practical adoption of NMT in code development. To that end, we develop a novel joint training approach for both localization and source code editions. Building a benchmark based on over 70k commits (patches and messages), we demonstrate that our jLED ( j oint L ocalize and ED it) approach is effective. An ablation study further demonstrates the importance of our design choice in joint training.

PENGABDIAN MASYARAKAT: PELATIHAN PENGGUNAAN MICROSOFT EXCEL PADA SISWA/I SD GMIT BELEMANA

Computer technology is a crucial aspect of daily life; however, many individuals, including students of SD GMIT Belemana, struggle to operate computers effectively. To address this issue, we conducted a Microsoft Excel training program aimed at enhancing students' abilities to adapt to technology. The training involved both theoretical explanations and hands-on practice. The activities included inputting data into columns, organizing tables neatly, and learning how to save their completed files. This training is expected to improve students' skills in utilizing technology, particularly Microsoft Excel, in the educational field, ultimately fostering a generation that can adapt to technological advancements and apply Microsoft Excel effectively. The method employed was a direct training approach, where we also introduced students to the various symbols used in Microsoft Excel. The results of this training showed that students experienced significant improvement in their ability to operate Microsoft Excel correctly and effectively, enabling them to apply the program in practical scenarios.

Evaluating YouTube’s Role in Records Management Training: Enhancing Knowledge and Job Performance for Records Personnel in Putrajaya

Abstract: This study explores the role of YouTube as a mediating variable between enhancing knowledge (independent variable) and job performance (dependent variable) for records personnel in Putrajaya. A total of 97 respondents (out of 118) from various public sector organizations participated in this study. Data was collected via a structured questionnaire, and statistical analysis was conducted using SPSS to examine descriptive statistics, reliability tests, and regression analysis. The results revealed a negative relationship between YouTube’s role in enhancing knowledge and job performance. Although YouTube was found to be useful for knowledge acquisition, its role did not significantly translate into improved job performance for records personnel. The study emphasizes the need for more tailored and formalized training approaches to bridge the gap between knowledge enhancement and job performance.

Eksplorasi Data Sains dalam Pendidikan: Membuka Wawasan Baru untuk Siswa SMA

Data science literacy is an important skill in the digital era, but high school students' understanding of this concept is still limited. This study aims to improve students' data science literacy through interactively designed training, including theory delivery and discussion sessions. This study was conducted in Surabaya involving 32 students from five partner schools. The research instruments included a pre-training questionnaire to measure the level of initial understanding, training materials in the form of presentations and educational case studies, and a post-training questionnaire to evaluate the results. The analysis was carried out quantitatively based on the results of the pre-test and post-test, and qualitatively through observation and discussion. The results showed a significant increase in student understanding, with the average post-test score increasing from 45% to 80%. Positive responses from students and teachers indicated that the case study-based training approach was relevant and effective. However, obstacles such as differences in student understanding levels and limited training duration were obstacles to implementation in maximizing knowledge for participants.

Best Practices in Cybersecurity Training: Enhancing Employee Awareness to Mitigate Risks

This technical article explores the critical role of cybersecurity training programs in addressing the growing challenges of cyber threats in modern organizations. The article examines how human factors contribute to security vulnerabilities while also presenting opportunities for strengthening organizational defense through comprehensive training initiatives. Through this detailed article of current research and industry reports, this article investigates the implementation of structured training approaches, measurement frameworks, and cultural transformation strategies. The article encompasses key aspects including baseline assessments, multi-tiered training architectures, performance metrics, security-conscious culture development, technical integration through learning management systems, compliance documentation, and future-proofing methodologies. The article demonstrates that organizations implementing comprehensive security awareness programs experience significant improvements in threat detection, incident response, and overall security posture while highlighting the importance of leadership engagement and continuous adaptation in maintaining effective cybersecurity training programs.

Effectiveness of integrated approach of emotional first aid and life skills training on psychological distress among adolescent students

Effectiveness of integrated approach of emotional first aid and life skills training on psychological distress among adolescent students

Improving spleen segmentation in ultrasound images using a hybrid deep learning framework

This paper introduces a novel method for spleen segmentation in ultrasound images, using a two-phase training approach. In the first phase, the SegFormerB0 network is trained to provide an initial segmentation. In the second phase, the network is further refined using the Pix2Pix structure, which enhances attention to details and corrects any erroneous or additional segments in the output. This hybrid method effectively combines the strengths of both SegFormer and Pix2Pix to produce highly accurate segmentation results. We have assembled the Spleenex dataset, consisting of 450 ultrasound images of the spleen, which is the first dataset of its kind in this field. Our method has been validated on this dataset, and the experimental results show that it outperforms existing state-of-the-art models. Specifically, our approach achieved a mean Intersection over Union (mIoU) of 94.17% and a mean Dice (mDice) score of 96.82%, surpassing models such as Splenomegaly Segmentation Network (SSNet), U-Net, and Variational autoencoder based methods. The proposed method also achieved a Mean Percentage Length Error (MPLE) of 3.64%, further demonstrating its accuracy. Furthermore, the proposed method has demonstrated strong performance even in the presence of noise in ultrasound images, highlighting its practical applicability in clinical environments.

Improving The Teachers Instructional Capacity through Assistance in Creating Microlearning-Based Learning Media

This community service activity aims to provide insight and knowledge of microlearning learning media for teachers in laboratory elementary schools. The training and mentoring approach is used in this community service activity. This community service activity is carried out through the stages of socialization, training, technology application, mentoring and evaluation, and program sustainability. At the evaluation stage, it was carried out with pre-test and post- test sheets which were then analyzed using the SPSS 24.0 program. Based on the results of post-test calculations, there was a significant improvement in the skills of participants before and after participating in training activities. Thus, it can be concluded that there is a need for regular training and mentoring activities, especially in increasing the instructional capacity of teachers in the current digital era. Through this community service activity, it is hoped that it will be able to produce a teaching module and post-activity monitoring.

Considerations for adapting digital competencies and training approaches to the public health workforce: an interpretive description of practitioners’ perspectives in Canada

BackgroundWidespread digital transformation necessitates developing digital competencies for public health practice. Given work in 2024 to update Canada’s public health core competencies, there are opportunities to consider digital competencies. In our previous research, we identified digital competency and training recommendations within the literature. In this study, we explored public health practitioners' experiences and perspectives on adapting identified digital competencies and training recommendations for Canada.MethodsBetween November and December 2023, we conducted an interpretive description using four focus groups with 19 public health practitioners working in regional and federal health authorities across Canada, with at least 3 years’ experience in current roles and experience using digital technologies in practice. We explored practitioners’ experiences using digital technologies and sought their opinions on how digital competency recommendations previously identified could be adapted to Canada’s context. To generate deep insights of practitioners’ subjective experiences and perspectives, we analyzed verbatim transcripts using Braun and Clarke’s reflexive thematic analysis.ResultsWe identified three main themes: a) public health systems must evolve to support new digital competencies; b) strengthen the basics before extending towards digital competencies; and c) focus on building general digital competencies with options for specialization where necessary. Findings emphasized matching workforce digital competencies to public health system capabilities and meaningfully integrating digital competencies within existing curricula. Such integration can consider how digital technologies change current public health practice to ensure practitioners are better able to address contemporary public health problems. Findings demonstrated roles for specialized digital programs as resources for learning within health systems and emphasized hands-on real-world training approaches.ConclusionWe need integrated, systems-focused approaches to digital competencies cutting across the current public health curriculum, while creating space for specialized digital public health competencies and roles. Further research is needed to understand requirements for enacting these recommendations in practice.

Peningkatan Pemahaman Guru Sekolah Indonesia Davao tentang Proyek Penguatan Profil Pelajar Pancasila Melalui Pendampingan Pembelajaran Terstruktur

This community service aims to increase the understanding of Indonesian Davao School Teachers (SID) in the Philippines regarding the Strengthening Pancasila Student Profile Project (P5). Through an In House Training approach based on structured learning, this activity succeeded in increasing teachers' understanding of P5 implementation. The mentoring activities were attended by 20 teachers at SID from elementary, junior and senior high school levels. The results of the mentoring show that there is an increase in teacher understanding regarding the implementation of P5 in the Kurikulum Merdeka. This is shown by the increase in the results of the teacher pre-test and post-test at SID amounting to 25.33 with an n-gain of 0.39 in the medium category. This activity also identified the need for more specific and focused training to support the optimization of the implementation of the Kurikulum Merdeka. As a follow-up, this service will be directed at developing advanced training modules and providing project-based assessment assistance. In addition, collaboration with various parties will be strengthened to ensure program sustainability and improve the quality of learning that is more inclusive and meaningful. This activity is a strategic step in supporting the transformation of Indonesian school education abroad.

A blended learning approach for capacity strengthening to improve the quality of integrated HIV, TB, and malaria services during antenatal and postnatal care in LMICs: a feasibility study

BackgroundThe blended learning (BL) approach to training health care professionals is increasingly adopted in many countries because of high costs and disruption to service delivery in the light of severe human resource shortage in low resource settings. The Covid-19 pandemic increased the urgency to identify alternatives to traditional face-to-face (f2f) education approach. A four-day f2f antenatal care (ANC) and postnatal care (PNC) continuous professional development course (CPD) was repackaged into a 3-part BL course; (1) self-directed learning (16 h) (2) facilitated virtual sessions (2.5 h over 3 days) and (3) 2-day f2f sessions. This study assessed the feasibility, change in healthcare providers’ knowledge and costs of the BL package in Nigeria, Tanzania, and Kenya.MethodsA mixed methods design was used. A total of 89 healthcare professionals, were purposively selected. Quantitative data was collected through an online questionnaire and skills assessments, analyzed using STATA 12 software. Qualitative data was collected through key informant interviews and focus group discussions, analysed using thematic analysis.ResultsMajority of participants (86%) accessed the online sessions using a mobile phone from home and health facilities. The median (IQR) time of completing the self-directed component was 16 h, IQR (8, 30). A multi-disciplinary team comprising of 42% nurse-midwives, 28% doctors, 20% clinical officers and 10% other healthcare professionals completed the BL course. Participants liked the BL approach due to its flexibility in learning, highly educative/relevant content, mixing of health worker cadres and CPD points. Aspects that were noted as challenging were related to personal log-in details and network connectivity issues during the self-directed learning and facilitated virtual sessions respectively.ConclusionThe blended learning approach to ANC-PNC in-service training was found to be acceptable, feasible and cost less to implement compared to face-to-face training approach in the study settings. The BL training approach was effective in improving the knowledge and skills of healthcare providers who participated in the training.

A generalist medical language model for disease diagnosis assistance.

The delivery of accurate diagnoses is crucial in healthcare and represents the gateway to appropriate and timely treatment. Although recent large language models (LLMs) have demonstrated impressive capabilities in few-shot or zero-shot learning, their effectiveness in clinical diagnosis remains unproven. Here we present MedFound, a generalist medical language model with 176 billion parameters, pre-trained on a large-scale corpus derived from diverse medical text and real-world clinical records. We further fine-tuned MedFound to learn physicians' inferential diagnosis with a self-bootstrapping strategy-based chain-of-thought approach and introduced a unified preference alignment framework to align it with standard clinical practice. Extensive experiments demonstrate that our medical LLM outperforms other baseline LLMs and specialized models in in-distribution (common diseases), out-of-distribution (external validation) and long-tailed distribution (rare diseases) scenarios across eight specialties. Further ablation studies indicate the effectiveness of key components in our medical LLM training approach. We conducted a comprehensive evaluation of the clinical applicability of LLMs for diagnosis involving artificial intelligence (AI) versus physician comparison, AI-assistance study and human evaluation framework. Our proposed framework incorporates eight clinical evaluation metrics, covering capabilities such as medical record summarization, diagnostic reasoning and risk management. Our findings demonstrate the model's feasibility in assisting physicians with disease diagnosis as part of the clinical workflow.