Training Model Research Articles

HANNES: A modular neurointerventional training model

Abstract This work introduces Version 2.0 of the Hamburg ANatomical NEurointerventional Simulator (HANNES): a novel modular neurointerventional simulator designed to provide a realistic platform for simulating various neurovascular pathologies and their endovascular therapies. Collaboratively developed by physicians and engineers, the neurointerventional simulator is designed with a modular product architecture combining standardised and variant modules to represent different training scenarios. The additively manufactured patient-based and patient-specific blood vessel tree consists of up to 13 individual components, including standardised features and interfaces for ease of connection. Patient-specific vessel models derived from medical imaging data allow customisation and complexity adjustment. HANNES supports diverse neurointerventional training scenarios, including various modalities of aneurysm embolisation, internal carotid artery stenosis, and thrombotic vessel occlusions, which can be treated through a transradial or transfemoral approach. The use and benefits of the model were evaluated with a group of trainees, who provided positive feedback, confirming the model's practicality and effectiveness in enhancing neurointerventional technical skills. In conclusion, HANNES represents a significant advancement in neurointerventional training, addressing the limitations of traditional training methods by simulating diverse disease patterns, enhancing medical staff's skills, and facilitating product testing.

Dimensional assessment on baseline MRI of soft-tissue sarcomas: longest diameter, sum and product of diameters, and volume-which is the best measurement method to predict patients' outcomes?

The longest diameter (LD) is a strong prognostic factor for patients with soft-tissue sarcoma (STS). Other dimensional assessments, such as the sum of diameters (SoD), product of diameters (PoD), and volume (3D-COG - proposed by the Children Oncology Group), can be rapidly performed; however, their prognostic values have never been compared to LD. Our goal was to investigate their performance in improving patients' prognostication for STS of the lower limbs. All consecutive adults managed with curative intent at our sarcoma reference center for a newly diagnosed STS of the lower limbs between 2000 and 2017, with pre-treatment MRI, were included in this retrospective study. Multivariable Cox regression models were trained to predict metastasis-free survival (MFS) in a Training cohort of 66.7% patients based on LD, PoD, SoD, or 3D-COG (and systematically including age, histologic grade, histotype, radiotherapy, chemotherapy, and surgical margins as covariables). The models were then compared on a validation cohort of 33.3% patients using concordance indices (c-index). The same approach was applied for overall survival (OS) and local relapse-free survival (LFS). Measurement reproducibility among three readers was evaluated with an intraclass correlation coefficient (ICC). 382 patients were included in the survival modeling (72/253 [28.5%] metastatic relapses in Training and 36/129 [27.9%] metastatic relapses in Validation). Higher dimensions were associated with lower MFS (multivariable hazard ratio [HR] = 2.44 and P = 0.0018 for LD; HR = 1.88 and P = 0.0009 for PoD, HR = 1.52 and P = 0.0041 for SoD; and HR = 1.08 and P = 0.0195 for 3D-COG). Higher c-indices were obtained with PoD model in Training (c-index = 0.772) and Validation (c-index = 0.688), but they were not significantly higher thanthose obtained with LD model. None of the measurements was associated with LFS or OS. All measurements demonstrated excellent ICC (> 0.95). Regarding its simplicity and good performance, LD appeared as the best metric to incorporate in prognostic models and nomograms for MFS.

Training deep learning models on personalized genomic sequences improves variant effect prediction.

Sequence-to-function models have broad applications in interpreting the molecular impact of genetic variation, yet have been criticized for poor performance in this task. Here we show that training models on functional genomic data with matched personal genomes improves their performance at variant effect prediction. Variant effect representations are retained even when transfer learning models to unseen cellular contexts and experimental readouts. Our results have implications for interpreting trait-associated genetic variation.

Framework to enable and test conversational assistant for APIs and RPAs

AbstractIn the realm of business automation, conversational assistants are emerging as the primary method for making automation software accessible to users in various business sectors. Access to automation primarily occurs through application programming interface (APIs) and robotic process automation (RPAs). To effectively convert APIs and RPAs into chatbots on a larger scale, it is crucial to establish an automated process for generating data and training models that can recognize user intentions, identify questions for conversational slot filling, and provide recommendations for subsequent actions. In this paper, we present a technique for enhancing and generating natural language conversational artifacts from API specifications using large language models (LLMs). The goal is to utilize LLMs in the “build” phase to assist humans in creating skills for digital assistants. As a result, the system does not need to rely on LLMs during conversations with business users, leading to efficient deployment. Along with enabling digital assistants, our system employs LLMs as proxies to simulate human interaction and automatically evaluate the digital assistant's performance. Experimental results highlight the effectiveness of our proposed approach. Our system is deployed in the IBM Watson Orchestrate product for general availability.

Neurosurgical Microvascular Anastomosis: Systematic Review of the Existing Simulators and Proposal of a New Training Classification System.

The literature lacks a combined analysis of neurosurgical microvascular anastomosis training models. We performed a systematic literature search to provide an overview of the existing models and proposed a classification system based on the level of simulation and reproducibility of the microvascular anastomosis. The systematic literature search followed the PRISMA guidelines. We consulted MEDLINE, Web of Knowledge, and EMBASE independently for papers about bypass training models. Every training model was analyzed according to six tasks supposed to esteem their fidelity to the real operative setting by using a scoring system from zero to two. Finally, authors classified the models into five classes, from A to E, by summing the individual scores. This study included 109 papers for analysis. Training models were grouped into synthetic tubes, ex vivo models (animal vessels, fresh human cadavers, human placentas) and in vivo simulators (live animals-rats, rabbits, pigs). By applying the proposed classification system, live animals and placentas obtained the highest scores, falling into class A (excellent simulators). Human cadavers and animal vessels (ex vivo) were categorized in class B (good simulators), followed by synthetic tubes (class C, reasonable simulators). The proposed classification system helps the neurosurgeon to analyze the available training models for microvascular anastomosis critically, and to choose the most appropriate one according to the skills they need to improve.

FLCMC: Federated Learning Approach for Chinese Medicinal Text Classification.

Addressing the privacy protection and data sharing issues in Chinese medical texts, this paper introduces a federated learning approach named FLCMC for Chinese medical text classification. The paper first discusses the data heterogeneity issue in federated language modeling. Then, it proposes two perturbed federated learning algorithms, FedPA and FedPAP, based on the self-attention mechanism. In these algorithms, the self-attention mechanism is incorporated within the model aggregation module, while a perturbation term, which measures the differences between the client and the server, is added to the local update module along with a customized PAdam optimizer. Secondly, to enable a fair comparison of algorithms' performance, existing federated algorithms are improved by integrating a customized Adam optimizer. Through experiments, this paper first conducts experimental analyses on hyperparameters, data heterogeneity, and validity on synthetic datasets, which proves that the proposed federated learning algorithm has significant advantages in classification performance and convergence stability when dealing with heterogeneous data. Then, the algorithm is applied to Chinese medical text datasets to verify its effectiveness on real datasets. The comparative analysis of algorithm performance and communication efficiency shows that the algorithm exhibits strong generalization ability on deep learning models for Chinese medical texts. As for the synthetic dataset, upon comparing with comparison algorithms FedAvg, FedProx, FedAtt, and their improved versions, the experimental results show that for data with general heterogeneity, both FedPA and FedPAP show significantly more accurate and stable convergence behavior. On the real Chinese medical dataset of doctor-patient conversations, IMCS-V2, with logistic regression and long short-term memory network as training models, the experiment results show that in comparison to the above three comparison algorithms and their improved versions, FedPA and FedPAP both possess the best accuracy performance and display significantly more stable and accurate convergence behavior, proving that the method in this paper has better classification effects for Chinese medical texts.

DDoS Attack Detection with Machine Learning

Nowadays, Distributed Denial of Service (DDoS) attacks are a major issue in internet security. These attacks target servers or network infrastructure. Similar to an unanticipated traffic jam on highway (lagging/crash) that prevent normal traffic reach to destination. DDoS may prevent users to access any system services. Researchers and scientists have developed numerous methods and algorithms to improve the performance of DDoS detection. In this paper, a DDoS detection method utilizing machine learning is proposed. There are three type of supervised machine learning classification methods which are K-Nearest Neighbor, Multilayer Perceptron and Random Forest, are applied in the proposed work to assess the accuracy of the model in training and testing processes. RF classification provides robustness and interpretability, MLP offers deep learning capabilities for complex patterns, and K-NN delivers simplicity and adaptability for instance-based learning. Together, these methods can contribute to a comprehensive DDoS attack detection system using machine learning. There are two types of classification setups: binary and multi-class classification. Binary classification involves identifying traffic as either a DDoS attack or normal using the NSL-KDD dataset. Multi-class classification, on the other hand, distinguishes between various types of DDoS attacks (such as DoS, Probe, U2R, and Sybil) and normal traffic using the NSL-KDD dataset. Feature engineering is also involved in this experiment to convert the categorical features into numerical values for detecting DDoS attack. Our model's performance was effective compared to other machine learning methods. RF achieved the highest accuracy rates: 99.35% in binary classification and 97.71% in multi-class classification. K-NN followed with 99.15% in binary and 97.35% in multi-class classification, while MLP achieved 90.63% in binary and 84.33% in multi-class classification.

Plastic Constitutive Training Method for Steel Based on a Recurrent Neural Network

The deep learning steel plastic constitutive model training method was studied based on the recurrent neural network (RNN) model to improve the allocative efficiency of the deep learning steel plastic constitutive model and promote its application in practical engineering. Two linear hardening constitutive datasets of steel were constructed using the Gaussian stochastic process. The RNN, long short-term memory (LSTM), and gated recurrent unit (GRU) were used as models for training. The effects of the data pre-processing method, neural network structure, and training method on the model training were analyzed. The prediction ability of the model for different scale series and the corresponding data demand were evaluated. The results show that LSTM and the GRU are more suitable for stress–strain prediction. The marginal effect of the stacked neural network depth and number gradually decreases, and the hysteresis curve can be accurately predicted by a two-layer RNN. The optimal structure of the two models is A50-100 and B150-150. The prediction accuracy of the models increased with the decrease in batch size and the increase in training batch, and the training time also increased significantly. The decay learning rate method could balance the prediction accuracy and training time, and the optimal initial learning rate, batch size, and training batch were 0.001, 60, and 100, respectively. The deep learning plastic constitutive model based on the optimal parameters can accurately predict the hysteresis curve of steel, and the prediction abilities of the GRU are 6.13, 6.7, and 3.3 times those of LSTM in short, medium, and long sequences, respectively.

4D printing of magnetoresponsive soft gripper and phenomenological approach for required magnetical actuation field

4D printing is the next step in additive manufacturing. Magnetoresponsive materials facilitate the creation of gripping tools through 4D printing, allowing for structural changes in response to external stimuli. In this study, the structural change is manifested as motion, triggered by an external magnetic field. This technology offers significant advantages in medical and industrial applications, including the printing of life-like moving organ models for medical training and the development of actuators for use in explosive environments. Magnetoresponsive materials are programmed with a magnetic profile and actuated by an external magnetic field. A compound of strontium ferrite microparticles SrFe12O19\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$SrFe_{12}O_{19}$$\\end{document}(≤20μm\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ \\le 20\\mu m$$\\end{document}) and an elastic polymer (thermoplastic copolyester) with a Hardness of Shore D 40 was produced. A star-shaped body was programmed and actuated by two permanent magnets, each of Br=1.29-1.32T\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$B_r=1.29 - 1.32T$$\\end{document}. As there is no analytical approach for calculating the required actuation flux density, one has been developed. The approach is verified experimentally by using a Hall probe. It is appropriate to set the field with a Helmholtz coil, despite the utilization of two permanent magnets. The use of a commercial fused filament fabrication printer for the processing of magnetoresponsive materials has been realized here for the first time. The main contributions are the short time constant (around ta=0.1s\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$t_a=0.1s$$\\end{document}) for actuation and the repeatability (around n=200\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$n=200$$\\end{document} actuation cycles) of the motion. The feasibility of multiple diverse reprogramming is a step forward in 4D printing. Hence, the post-print programming and the inhomogeneity of the field limit the ease of the presented method.

Method for Enhancing AI Accuracy in Pressure Injury Detection Using Real and Synthetic Datasets

Pressure injuries pose significant health risks, especially for the elderly, immobile individuals, and those with sensory impairments. These injuries can rapidly become chronic, making initial diagnosis important. Due to the difficulty of transporting patients from local health facilities to higher-level general hospitals for treatment, it is essential to utilize telemedicine tools, such as chatbots, to ensure rapid initial diagnosis. Recent advances in artificial intelligence have demonstrated potential for medical imaging and disease classification. Ongoing research in the field of dermatological diseases focuses on disease classification. However, the assessment accuracy of artificial intelligence is often limited by unequal class distributions and insufficient dataset quantities. In this study, we aim to enhance the accuracy of artificial intelligence models by generating synthetic datasets. Specifically, we focused on training models for Pressure Injury assessment using both real and synthetic datasets. We used PI data at a domestic medical university. As part of our supplementary research, we established a chatbot system to facilitate the assessment of pressure injuries. Using both constructed and synthetic data, we achieved a top-1 accuracy of 92.03%. The experimental results demonstrate that combining real and synthetic data significantly improves model accuracy. These findings suggest that synthetic datasets can be effectively utilized to address the limitations of small-scale datasets in medical applications. Future research should explore the use of diverse synthetic data generation methods and validate model performance on a variety of datasets to enhance the generalization and robustness of AI models for Pressure Injury assessment.

Computer vision-based algorithm for precise defect detection and classification in photovoltaic modules

In recent years, driven by advancements in the photovoltaic industry, solar power generation has emerged as a crucial energy source in China and the globe. A progressive annotation approach is employed to pinpoint and label defect samples to enhance the precision of automated detection technology for minor defects within photovoltaic modules. Subsequently, computer vision techniques are harnessed to segment photovoltaic modules and defect samples amidst intricate backgrounds accurately. Finally, a transfer learning training model is deployed to classify and identify defects effectively. The results indicate that the mask-region convolutional neural network model achieves remarkable accuracy and recall rates of 98.7% and 0.913, respectively. Furthermore, the detection speed and inference time are 280.69 frames per second and 3.53 ms, respectively. In essence, the defect detection and classification algorithm utilizing computer vision techniques significantly enhances the precision of automated detection technology in identifying minor defects within complex environments. This advancement holds profound practical significance in ensuring photovoltaic modules’ quality and operational reliability.

The Integration of Nurse Training into the University of Swaziland and its Impact on the Professionalisation of Nursing, 1980s–2007

The turn of the twentieth century witnessed the onset of a discussion about how to properly educate a nurse. This discussion highlighted the irrelevance of the hospital-based nurse training model, ultimately pushing nurse training to colleges associated with universities. Scholarship on the transition from hospital to university-based nurse training in southern Africa is thin. In Swaziland (now Eswatini), the affiliation of nursing colleges with the University of Swaziland (UNISWA) occurred in 1989. Integrating nurse training into UNISWA was an impetus for major shifts in nurse education in the territory, one of which was the professionalisation of nursing as a career. Profound as this development was, nothing has been written about it. Drawing from a wide range of sources, this article argues that the integration of nurse training into UNISWA was a desirable transformation in nurse education which enhanced the quality of nurse care service provision and professionalised nursing. The article provides the contextual background prior to the integration of nurse training into UNISWA for us to understand how nurse training was carried out in the period up to 1989. It discusses factors which facilitated the transition. It also examines this transition and explores its impact on the professionalisation of nurse training in Swaziland.

Development and Validation of a Laparoscopic Sacrocolpopexy Training Model.

Proper training is necessary to develop the highly specialized skills required to safely perform laparoscopic sacrocolpopexy. Currently, there is no validated training model for laparoscopic sacrocolpopexy that includes dissection of the presacral space, both vaginal and presacral mesh attachments, and peritoneal closure. This study aimed to create a procedure specific hierarchical task analysis for laparoscopic sacrocolpopexy and then develop and validate a corresponding laparoscopic sacrocolpopexy pelvic training model for the simulation environment. This was an observational simulation study that was divided into 5 phases: (1) development of hierarchical task analysis, (2) model construction, (3) participant recruitment and simulation testing, (4) reliability and validity testing, and (5) creation of a standard passing performance measure. Construct, face, and content validity were established for this model. According to the participating experts, the model was able to replicate the steps of presacral dissection, anterior vaginal and sacral mesh attachment, and peritoneal closure. Thirteen trainees and 5 experts completed the simulation, and all "agreed" or "strongly agreed" that the model seemed useful for improving suturing technique and learning the procedure. Additionally, a passing performance measure was determined through contrasting groups methodology. We developed a novel, reusable, and validated training model that can be utilized as a training resource for the many critical skills necessary to safely and efficiently perform laparoscopic sacrocolpopexy.

Retrieval of Solar Shortwave Irradiance from All-Sky Camera Images

The present work proposes a new model based on a convolutional neural network (CNN) to retrieve solar shortwave (SW) irradiance via the estimation of the cloud modification factor (CMF) from daytime sky images captured by all-sky cameras; this model is named CNN-CMF. To this end, a total of 237,669 sky images paired with SW irradiance measurements obtained by using pyranometers were selected at the following three sites: Valladolid and Izaña, Spain, and Lindenberg, Germany. This dataset was randomly split into training and testing sets, with the latter excluded from the training model in order to validate it using the same locations. Subsequently, the test dataset was compared with the corresponding SW irradiance measurements obtained by the pyranometers in scatter density plots. The linear fit shows a high determination coefficient (R2) of 0.99. Statistical analyses based on the mean bias error (MBE) values and the standard deviation (SD) of the SW irradiance differences yield results close to −2% and 9%, respectively. The MBE indicates a slight underestimation of the CNN-CMF model compared to the measurement values. After its validation, model performance was evaluated at the Antarctic station of Marambio (Argentina), a location not used in the training process. A similar comparison between the model-predicted SW irradiance and pyranometer measurements yielded R2=0.95, with an MBE of around 2% and an SD of approximately 26%. Although the precision provided by the SD at the Marambio station is lower, the MBE shows that the model’s accuracy is similar to previous results but with a slight overestimation of the SW irradiance. Finally, the determination coefficient improved to 0.99, and the MBE and SD are about 3% and 11%, respectively, when the CNN-CMF model is used to estimate daily SW irradiation values.

Improving rail transit security with enhanced YOLOv5 obstacle detection

Abstract As the economy advances and rail transit technology progresses, rail transit plays an important role in alleviating urban traffic pressure and promoting regional economic development. However, railway safety faces potential hazards such as foreign object intrusion in the trackside environment, seriously threatening the safety of trains and passengers. Therefore, to achieve high-performance recognition of obstacle target images in rail transit, a small sample object detection algorithm YOLOv5-RTO, which using EVC (Explicit Visual Center) attention mechanism and lightweight up-sampling operator CARAFE (Content-Aware ReAssembly of FEatures) to improve YOLOv5, has been proposed. Based on the self-collected image data of rail transit obstacles, a dense and time-varying dataset was created within the rail transit system, and data augmentation methods were used to reduce the impact of small sample datasets on the detection performance of subsequent algorithm. Using this dataset, target detection tests were conducted on rail transit obstacle images to analyze the actual performance of the training model. The experiment results indicate that the enhanced YOLOv5-RTO algorithm integrates the advantages of EVC and CARAFE while preserving the lightweight and accurate characteristics of YOLOv5. The improved algorithm scheme has an average inference time of 1.3 ms per frame in the field of rail transit images, achieving an overall mAP of 96.5%, an increase of 1.8% compared to the original algorithm. It has a certain engineering application value and provides reference for further optimizing the safety system of urban rail transit.

Who Gets Paid (for) What? The Cultural Political Economy of News Content in Generative AI

One of the key controversies that generative artificial intelligence (AI) has recently stirred was whether compensation is due for the copyrighted materials used to train AI models. This article explores the logic, trajectories, and dynamics of content generation, including news, through generative AI in two distinctive yet intertwined domains. Guided by a cultural political economy approach, it examines how both the political context (validation/legitimation of AI-generated news content by established news media) and the economic context (use of unpaid and underpaid labor in the forms of freely scraped data and data annotation work) shape the deployment of news content on AI models. It further untangles how the space for serious, independent journalism may shrink, as big tech companies’ algorithmic technologies emerge as a solution to contemporary problems in journalism. A clear danger here is that AI companies’ proprietary algorithms, language training models, and value-laden parameters are incompatible with journalism's democratic obligations and responsibilities.

Saber-toothing the University In–service Curriculum for Zimbabwean Primary Schools: The Contestations

In this study, the authors argue that ‘saber–toothing’ the university in-service curriculum is a critical success factor in building an impact educational programme for Zimbabwean primary schools. By saber-toothing the in–service curriculum, the authors propose that the education curriculum at university should continue to develop in order to address and serve the purposes and needs of the time. It explains how some unexamined habits and practices can result in resisting the much-needed curriculum reforms. The new global and national conditions demand a paradigm shift to chat new in–service curriculum discourses that develop the Zimbabwean education system. The study uses an interpretivist qualitative case study of one Zimbabwean university, purposively sampled five experienced and senior lecturers teaching the in–service programme and ten in–service student teachers. Data were collected through semi–structured interviews in respect of lecturers, focus group discussions within–service students and document review of module outlines. The main finding is that the in–service curriculum has remained stagnant in the changing times and failed in its highest end to develop a reflective and responsive in–service product that can ‘saber–tooth’ the knowledge and skills to solve challenges and realities of their time and contexts. The study recommends that universities re-vision the largely existing in-service academic curricula and practices to adopt a practically–oriented, knowledge generation training and skills-based model that equips teachers with the proper knowledge and skills to change educational fortunes in their unique local school contexts.

Intelligent agricultural robotic detection system for greenhouse tomato leaf diseases using soft computing techniques and deep learning

The development of soft computing methods has had a significant influence on the subject of autonomous intelligent agriculture. This paper offers a system for autonomous greenhouse navigation that employs a fuzzy control algorithm and a deep learning-based disease classification model for tomato plants, identifying illnesses using photos of tomato leaves. The primary novelty in this study is the introduction of an upgraded Deep Convolutional Generative Adversarial Network (DCGAN) that creates augmented pictures of disease tomato leaves from original genuine samples, considerably enhancing the training dataset. To find the optimum training model, four deep learning networks (VGG19, Inception-v3, DenseNet-201, and ResNet-152) were carefully compared on a dataset of nine tomato leaf disease classes. These models have validation accuracy of 92.32%, 90.83%, 96.61%, and 97.07%, respectively, when using the original PlantVillage dataset. The system then uses an enhanced dataset with ResNet-152 network design to achieve a high accuracy of 99.69%, as compared to the original dataset with ResNet-152’s accuracy of 97.07%. This improvement indicates the use of the proposed DCGAN in improving the performance of the deep learning model for greenhouse plant monitoring and disease detection. Furthermore, the proposed approach may have a broader use in various agricultural scenarios, potentially altering the field of autonomous intelligent agriculture.

The influence of a coordination training model that uses various manipulative movements on futsal kick accuracy

Achieving precise kick accuracy and coordinated movements in futsal was essential for competitive success. This study aimed to evaluate the differences between various futsal training approaches that incorporated distinct coordination exercises involving manipulative movements, specifically assessing their impact on kick accuracy. A quasi-experimental design with a control group was employed, focusing on shooting skills within the futsal context. The population consisted of 70 members from the Universitas Muhammadiyah Kalimantan Timur (UMKT) futsal team, from which 40 players were purposively selected based on their participation duration of over six months and their tier status within the university's core teams. Participants were divided into two groups: one received static coordination training (control group), and the other underwent dynamic coordination training (experimental group), each comprising 20 players. The intervention spanned 18 sessions, including pre- and post-tests. Data were collected through a modified shooting kick test at the futsal goal, utilizing a point distribution system. Analysis revealed significant differences in kick performance between the two training methods, with dynamic coordination training yielding superior results (t-count = 3.010 t-table = 2.024). These findings indicated that dynamic coordination exercises were more effective than static ones in enhancing futsal kick accuracy. Future research should have considered expanding the population and sample size to include amateur players, introducing additional variables and diverse training methodologies, extending training duration and frequency, conducting long-term evaluations, measuring shooting technique quality, and employing quantitative research methods. By addressing these suggestions, subsequent studies could contribute more comprehensively to the understanding of effective training strategies in futsal.

Research on the classroom teaching design model of enterprise mentors from the perspective of modern apprenticeship

With the advancement of modern apprenticeship, corporate mentors play an increasingly important role in vocational education. However, in the face of new talent training models, corporate mentors face many problems and challenges in classroom teaching design work. This study aims to investigate the problems encountered in the classroom teaching design process for corporate mentors from the perspective of modern apprenticeship. It proposes a classroom teaching design model based on the BOPPPS model that integrates digital technology to solve these problems. This model aims to enhance corporate mentors' teaching design capabilities and classroom instruction quality through digital technology empowerment and optimization of the BOPPPS model. The research results of the paper provide a theoretical, systematic, and scientific classroom teaching design model for enterprise instructors. It provides new perspectives and practical solutions to the classroom instructional design of enterprise instructors under the modern apprenticeship system.