Reality Training Research Articles

Accelerated Pure Shift NMR Spectroscopy with Deep Learning.

Pure shift nuclear magnetic resonance (NMR) spectroscopy presents a promising solution to provide sufficient spectral resolution and has been increasingly applied in various branches of chemistry, but the optimal resolution is generally accompanied by long experimental times. We present a proof of concept of deep learning for fast, high-quality, and reliable pure shift NMR reconstruction. The deep learning (DL) protocol allows one to eliminate undersampling artifacts, distinguish peaks with close chemical shifts, and reconstruct high-resolution pure shift NMR spectroscopy along with accelerated acquisition. More meaningfully, the lightweight neural network delivers satisfactory reconstruction performance on personal computers by several hundred simulated data learning, which somewhat lifts the prohibiting demand for a large volume of real training samples and advanced computing hardware generally required in DL projects. Additionally, an M-to-S strategy applicable to common DL cases is further exploited to boost the network generalization capability. As a result, this study takes a meaningful step toward deep learning protocols for broad chemical applications.

Virtual reality applications in pediatric surgery

Virtual reality modeling (VRM) is a 3-dimensional (3D) simulation. It is a powerful tool and has multiple uses and applications in pediatric surgery. Patient-specific 2-dimensional imaging can be used to generate a virtual reality model, which can improve anatomical perception and understanding, and can aid in preoperative planning for complex operations. VRM can also be used for realistic training and simulation. It has also proven effective in distraction for pediatric patients experiencing pain and/or anxiety. We detail the technical requirements and process required for VRM generation, the applications, and future directions.

Do extreme temperatures affect cognition? A short review of the impact of acute heat stress on cognitive performance of firefighters.

Research shows that exposure to high environmental temperatures can affect task performance. Theoretical explanations outline that heat is a source of stress that competes for limited-capacity resources, therefore if a task is resource-intensive, and/or if heat stress is extreme, performance will suffer. One occupation in which individuals complete demanding tasks and make difficult decisions, often in temperatures exceeding 200°C, is firefighting. Yet very little is currently known about the impact of heat stress on the cognitive functioning of firefighters. This short review summarizes the limited research in this area, focusing on studies that measured cognition of firefighters following a realistic training exercise. The findings are mixed with evidence that heat stress improves, impairs, and has no impact on cognitive functioning. While there are differences in the firefighting activities utilized, and the temperatures that participants were exposed to, it is argued that the varied findings can be attributed to the tasks used to assess cognitive processing, and the cognitive functions being measured. In accordance with the wider field of research, it is concluded that complex functioning, such as sustained attention, vigilance, and working memory is negatively impacted by acute exposure to extreme heat. Greater understanding of factors affecting cognition would inform safety practices and more research is needed to understand how and when heat stress may influence cognition in firefighting scenarios.

Synthetic Data and Hierarchical Object Detection in Overhead Imagery

The performance of neural network models is often limited by the availability of big datasets. To treat this problem, we survey and develop novel synthetic data generation and augmentation techniques for enhancing low/zero-sample learning in satellite imagery. In addition to extending synthetic data generation approaches, we propose a hierarchical detection architecture to improve the utility of synthetic training samples. We consider existing techniques for producing synthetic imagery–3D models and neural style transfer–as well as introducing our own adversarially trained reskinning network, the GAN-Reskinner, to blend 3D models. Additionally, we test the value of synthetic data in a two-stage, hierarchical detection/classification model of our own construction. To test the effectiveness of synthetic imagery, we employ it in the training of detection models and our two stage model, and evaluate the resulting models on real satellite images. All modalities of synthetic data are tested extensively on practical, geospatial analysis problems. Our experiments show that synthetic data developed using our approach can often enhance detection performance, particularly when combined with some real training images. When the only source of data is synthetic, our GAN-Reskinner often boosts performance over conventionally rendered 3D models and in all cases, the hierarchical model outperforms the baseline end-to-end detection architecture.

Application of Simulation Technology in Football Training: A Systematic Review of Empirical Studies

Background This review aimed to cover the characteristics and functions of simulation tools applied to football training, the process and results of empirical research, the benefits of simulation tools for football training, and existing challenges. Materials and Methods To investigate and analyze the effect of simulation technology in football training, the PRISMA method was used to systematically review 18 relevant empirical studies published between January, 2014 and July, 2023. Results The study identified three types of tools for applying simulation technology to football training, including head-mounted displays, Cave Automatic Virtual Environment (CAVE), and Screen-Based Simulation. These tools have been effective in training football playing techniques (including goalkeeping, heading, etc.) as well as football tactical skills (including perception-cognitive and decision-making) and can be used as a supplement to regular training. Conclusion If simulation technology is to enhance football training, we suggest that it is necessary to carefully verify the validity of the tool and the long-term impact of simulation training on participants and verify that simulation actually translates to real-world games. At the same time, it is suggested that future research could explore training with mixed VR and AR to develop more realistic and effective training platforms.

The Be REAL Framework: Enhancing Relationship-Building Skills for Community Health Workers.

Trust plays an integral part in the effective functioning of public health systems. During the COVID-19 pandemic, distrust of public health fueled vaccine hesitancy and created additional barriers to immunization. Although most Americans have received at least one COVID-19 vaccine, the percentage of fully immunized adults remains suboptimal. To reach vaccine-hesitant communities, it is vital that public health be worthy of trust. As trusted members of their communities, community health workers (CHWs) can serve as ideal messengers and conversation partners for vaccination decision-making. We developed the Be REAL framework and training materials to prepare CHWs to work with vaccine-hesitant communities nationwide. Through the four steps of "Relate," "Explore," "Assist," and "Leave (the door open)," CHWs were taught to prioritize relationship building as a primary goal. In this shift from focusing on adherence to public health recommendations (e.g., get vaccinated) to building relationships, the value of vaccine uptake is secondary to the quality of the relationship being formed. The Be REAL framework facilitates CHWs harnessing the power they already possess. The goal of the Be REAL framework is to foster true partnership between CHWs and community members, which in turn can help increase trust in the broader public health system beyond adherence to a specific recommendation.

Developing the Spirit of Entrepreneurship through Training in Making Sprinkle Chili Powder (Bon Cabe)


 
 
 
 Entrepreneurship is a constantly changing field, and people are always looking for chances to start businesses that will not only help them reach their own goals but also grow their communities' economies. One way is to make Sprinkle Chili Powder, which is also known as Bon Cabe. Many people love this versatile and well-known condiment, which makes it a great business chance for people who want to start their own business. To use this potential, a full training program has been created to encourage people to be entrepreneurial and teach them how to make high-quality Bon Cabe. Consumers will tend to choose products with attractive packaging, compared to products whose packaging is plain or simple and unattractive. There are quite a lot of chilli products produced in the Sidekarya Farmer Group. The process of processing chilli products into chilli bonds produced by partners is very constrained, especially in product processing, packaging and marketing according to prices. The main problem that will be the focus of solving the problem is how to make processed chillies into BonCabe (Sprinkle Chili Powder) and improve the product packaging so that it can attract consumers to buy products, Participants who want to get into the lucrative business of Bon Cabe making can use this training program as a stepping stone. Participants will not only learn the skills they need to be great entrepreneurs, but they will also help the local economy grow and become more diverse by combining classroom learning with hands-on experience. The program shows how real training can change the lives of people who want to become entrepreneurs and help them reach their full potential. It's not just about producing a well-liked condiment; it's also a life-changing event that enables people to turn their dreams into successful businesses.
 
 
 

The Benign Brand Operation of Professional Sports Leagues: A Case Study on Real Madrid

Real Madrid's profitability continued to grow during the epidemic period, making it a rare club in Europe to still be profitable during COVID-19. Based on this event, the purpose of this exploration was to exploare what kind of management style and business strategy Real Madrid used during the epidemic period of 20192022. Using Real Madrid as a case study, this paper analyzed the information sources applied to Real Madrid's revenues in different aspects as well as Real Madrid's business formula. The first reason was Real Madrid's market management philosophy, and the corresponding recommendation was to increase fan loyalty by strengthening the development of the players and improving the rights of the fans, along with digital marketing tools. The second reason was Real Madrid's excellent youth training system, and the corresponding recommendation was to make more use of modern data analysis systems to track the development of each player. The third reason was Real Madrid's team culture, and the recommendation was to utilize the celebrity effect to promote the team's culture. The main objective of such an exploration was to help national soccer clubs learn from Real Madrid's successful management model to improve their profitability.

Accurate Robot Arm Attitude Estimation Based on Multi-View Images and Super-Resolution Keypoint Detection Networks.

Robot arm monitoring is often required in intelligent industrial scenarios. A two-stage method for robot arm attitude estimation based on multi-view images is proposed. In the first stage, a super-resolution keypoint detection network (SRKDNet) is proposed. The SRKDNet incorporates a subpixel convolution module in the backbone neural network, which can output high-resolution heatmaps for keypoint detection without significantly increasing the computational resource consumption. Efficient virtual and real sampling and SRKDNet training methods are put forward. The SRKDNet is trained with generated virtual data and fine-tuned with real sample data. This method decreases the time and manpower consumed in collecting data in real scenarios and achieves a better generalization effect on real data. A coarse-to-fine dual-SRKDNet detection mechanism is proposed and verified. Full-view and close-up dual SRKDNets are executed to first detect the keypoints and then refine the results. The keypoint detection accuracy, PCK@0.15, for the real robot arm reaches up to 96.07%. In the second stage, an equation system, involving the camera imaging model, the robot arm kinematic model and keypoints with different confidence values, is established to solve the unknown rotation angles of the joints. The proposed confidence-based keypoint screening scheme makes full use of the information redundancy of multi-view images to ensure attitude estimation accuracy. Experiments on a real UR10 robot arm under three views demonstrate that the average estimation error of the joint angles is 0.53 degrees, which is superior to that achieved with the comparison methods.

Real needle for minimal invasive procedures training using motion sensors and optical flow

Minimally invasive percutaneous insertion procedures are widely used techniques in medicine. Their success is highly dependent on the skills of the practitioner. This paper presents a haptic simulator for training in these procedures, whose key component is a real percutaneous insertion needle with a sensory system incorporated to track its 3D location at every instant.By means of the proposed embedded vision system, the attitude (spatial orientation) and depth of insertion of a real needle are estimated. The proposal is founded on a novel depth estimation procedure based on optical flow techniques, complemented by sensory fusion techniques with the attitude calculated with data from an Inertial Measurement Unit (IMU) sensor. This procedure allows estimating the needle attitude with an accuracy of tenths of a degree and the displacement with an accuracy of millimeters. The computational algorithm runs on an embedded computer with real-time constraints for tracking the movement of a real needle. This haptic needle location data is used to reproduce the movement of a virtual needle within a simulation app.As a fundamental result, an ergonomic and realistic training simulator has been successfully constructed for healthcare professionals to acquire the mental model and motor skills necessary to practice percutaneous procedures successfully.

Factors Impeding Learning at Various Stages of Simulation Training as Experienced by Nursing Students

Aim: To identify the learning barriers, nursing students encounter throughout distinct stages of simulation‐based education.Design: Qualitative study exploring learning barriers in simulation training.Methods: Data were collected through interviews with 13 fourth‐year nursing students at Dongnam Health University, divided into four groups. These students had prior simulation experience with a variety of medical scenarios, including shortness of breath, chest pain, preterm labor, postoperative pain management, and neonatal care. Interviews were conducted using a semistructured format and lasted approximately one hour per group. We then used content analysis to identify primary and secondary themes.Results: Three main themes and 13 subthemes were identified, highlighting learning challenges across different stages of simulation training: limitations in integrating theoretical foundations during the prebriefing phase, the impact of psychological stress on clinical reasoning abilities during the scenario phase, and constraints in reflective thinking during the debriefing phase.Conclusion: This paper highlights significant learning obstacles across the simulation training process and advocates for targeted educational strategies to overcome these barriers. Specific recommendations include implementing additional measures to enhance student immersion, ensuring psychological safety, and enhancing debriefing sessions to encourage reflective thinking.Implications: This paper emphasizes the need for immersive and realistic training in nursing education. It suggests enhancements in applying theoretical knowledge, managing stress‐induced clinical reasoning, and fostering reflective thinking to better prepare students for clinical realities. It also acknowledges that the sample size may limit generalizability and recommends further studies with larger cohorts.Impact: This study aims to better prepare nursing students for the complexities of clinical practice by emphasizing the development of holistic patient care and adaptive problem‐solving skills. The research implications suggest a potential transformation in nursing education and practice, aiming to improve healthcare outcomes.

Harnessing augmented reality technology for medical education – a virtual abdominal hysterectomy

With the advent of augmented reality (AR), our team developed a prototype teaching module using the abdominal hysterectomy as a case study for the teaching of in-depth pelvic anatomy and the steps for performing a hysterectomy. Through this interactive module, trainees will be able to identify and fill knowledge gaps in pelvic anatomy with the benefit of spatial visualisation, and familiarise themselves with the steps of performing a hysterectomy in a safe controlled setting. The prototype teaching module has eight sections, each with its specific aims, objectives and multiple choice questions to assess the participant’s knowledge and understanding. The scores will be tabulated and a final report will be generated upon completion of the entire module. This includes the breakdown of the scores, mistakes identified and recommended suggestions to facilitate learning. Following completion of the prototype teaching module, we aim to carry out a trial on our residents and assess its effects on self-perceived preparedness, knowledge and impact on training. We believe that this can translate to better awareness with reduced errors in surgery and ultimately better patient outcomes and safety. Our new prototype teaching module promises to be well-received, relevant and will take our training to a new level. There is enormous potential in using AR technology to provide realistic, interactive and safe training and this can be harnessed for the future development of other modules in the O&G curriculum.

Research on intelligent forecasts of flight actions based on the implemented bi-LSTM.

Rapid identification of flight actions by utilizing flight data is more realistic so the quality of flight training can be objectively assessed. The bidirectional long short-term memory (bi-LSTM) algorithm is implemented to forecast the flight actions of aircraft. The dataset containing the flight actions is structured by collecting tagged flight data when real flight training is exercised. However, the dataset needs to be preprocessed and annotated with expert rules. One of the deep learning (DL) methods, called the bi-LSTM algorithm, is implemented to train and test, and the pivotal parameters of the algorithm are optimized. Finally, the constructed model is applied to forecast the flight actions of aircraft. The training's accuracy and loss rates are computed. The duration is kept between 1 through 3 h per session. Thus, the development of training the model is continued until an accuracy rate above 85% is achieved. The word-run inference time is kept under 2 s. Finally, the proposed algorithm's specific characteristics, which are short training time and high recognition accuracy, are achieved when complex rules and large sample sizes exist.

OpenGAN: Open-Set Recognition Via Open Data Generation.

Real-world machine learning systems need to analyze test data that may differ from training data. In K-way classification, this is crisply formulated as open-set recognition, core to which is the ability to discriminate open-set data outside the K closed-set classes. Two conceptually elegant ideas for open-set discrimination are: 1) discriminatively learning an open-vs-closed binary discriminator by exploiting some outlier data as the open-set, and 2) unsupervised learning the closed-set data distribution with a GAN, using its discriminator as the open-set likelihood function. However, the former generalizes poorly to diverse open test data due to overfitting to the training outliers, which are unlikely to exhaustively span the open-world. The latter does not work well, presumably due to the unstable training of GANs. Motivated by the above, we propose OpenGAN, which addresses the limitation of each approach by combining them with several technical insights. First, we show that a carefully selected GAN-discriminator on some real outlier data already achieves the state-of-the-art. Second, we augment the available set of real open training examples with adversarially synthesized "fake" data. Third and most importantly, we build the discriminator over the features computed by the closed-world K-way networks. This allows OpenGAN to be implemented via a lightweight discriminator head built on top of an existing K-way network. Extensive experiments show that OpenGAN significantly outperforms prior open-set methods.

MovePort: Multimodal Dataset of EMG, IMU, MoCap, and Insole Pressure for Analyzing Abnormal Movements and Postures in Rehabilitation Training.

In most real world rehabilitation training, patients are trained to regain motion capabilities with the aid of functional/epidural electrical stimulation (FES/EES), under the support of gravity-assist systems to prevent falls. However, the lack of motion analysis dataset designed specifically for rehabilitation-related applications largely limits the conduct of pilot research. We provide an open access dataset, consisting of multimodal data collected via 16 electromyography (EMG) sensors, 6 inertial measurement unit (IMU) sensors, and 230 insole pressure sensors (IPS) per foot, together with a 26-sensor motion capture system, under different MOVEments and POstures for Rehabilitation Training (MovePort). Data were collected under diverse experimental paradigms. Twenty four participants first imitated multiple normal and abnormal body postures including (1) normal standing still, (2) leaning forward, (3) leaning back, and (4) half-squat, which in practical applications, can be detected as feedback to tune the parameters of FES/EES and gravity-assist systems to keep patients in a target body posture. Data under imitated abnormal gaits, e.g., (1) with legs raised higher under excessive electrical stimulation, and (2) with dragging legs under insufficient stimulation, were also collected. Data under normal gaits with low, medium and high speeds are also included. Pathological gait data from a subject with spastic paraplegia further increases the clinical value of our dataset. We also provide source codes to perform both intra- and inter-participant motion analyses of our dataset. We expect our dataset can provide a unique platform to promote collaboration among neurorehabilitation engineers.

Denoising diffusion probabilistic models for addressing data limitations in chest X-ray classification

Deep learning plays a crucial role in medical imaging analysis, particularly in tasks such as image classification and segmentation. However, learning from medical imaging datasets presents challenges, including scarcity of labeled examples, class imbalances, and inadequate representation of diverse patient populations. To address these challenges, there has been a growing interest in the use of deep generative models to create synthetic training data, with denoising diffusion probabilistic models (DDPMs) recently gaining attention for their ability to produce realistic and high-quality images. This study explores the potential of a DDPM to generate synthetic chest X-rays for multi-label classifier training. The results indicate that the use of a conditional DDPM has the potential to produce a realistic training set of synthetic chest X-rays. In addition, the study analyzes the impact on classification performance of addressing class imbalance. Balancing the synthetic training set increased the overall classification sensitivity from 0.02 to 0.59, but decreased the overall specificity from 0.99 to 0.71. Furthermore, we investigated the potential of unconditional pre-training to learn general representations, followed by conditional fine-tuning of the DDPM. The results indicate that this approach allows the amount of labeled training data to be reduced to 25% of the original set. Finally, we demonstrate that fidelity and classification metrics do not consistently exhibit the same trends. Integrating a DDPM into the classification pipeline underscores the benefits of having optimal control over the data and efficient use of available unlabeled data. Our research provides insights for making informed decisions about integrating generative models into medical image analysis.

CPPF++: Uncertainty-Aware Sim2Real Object Pose Estimation by Vote Aggregation.

Object pose estimation constitutes a critical area within the domain of 3D vision. While contemporary state-of-the-art methods that leverage real-world pose annotations have demonstrated commendable performance, the procurement of such real training data incurs substantial costs. This paper focuses on a specific setting wherein only 3D CAD models are utilized as a priori knowledge, devoid of any background or clutter information. We introduce a novel method, CPPF++, designed for sim-to-real category-level pose estimation. This method builds upon the foundational point-pair voting scheme of CPPF, reformulating it through a probabilistic view. To address the challenge posed by vote collision, we propose a novel approach that involves modeling the voting uncertainty by estimating the probabilistic distribution of each point pair within the canonical space. Furthermore, we augment the contextual information provided by each voting unit through the introduction of N-point tuples. To enhance the robustness and accuracy of the model, we incorporate several innovative modules, including noisy pair filtering, online alignment optimization, and a tuple feature ensemble. Alongside these methodological advancements, we introduce a new category-level pose estimation dataset, named DiversePose 300. Empirical evidence demonstrates that our method significantly surpasses previous sim-to-real approaches and achieves comparable or superior performance on novel datasets. Our code is available on https://github.com/qq456cvb/CPPF2.

FocalPose++: Focal Length and Object Pose Estimation Via Render and Compare.

We introduce FocalPose++, a neural render-and-compare method for jointly estimating the camera-object 6D pose and camera focal length given a single RGB input image depicting a known object. The contributions of this work are threefold. First, we derive a focal length update rule that extends an existing state-of-the-art render-and-compare 6D pose estimator to address the joint estimation task. Second, we investigate several different loss functions for jointly estimating the object pose and focal length. We find that a combination of direct focal length regression with a reprojection loss disentangling the contribution of translation, rotation, and focal length leads to improved results. Third, we explore the effect of different synthetic training data on the performance of our method. Specifically, we investigate different distributions used for sampling object's 6D pose and camera's focal length when rendering the synthetic images, and show that parametric distribution fitted on real training data works the best. We show results on three challenging benchmark datasets that depict known 3D models in uncontrolled settings. We demonstrate that our focal length and 6D pose estimates have lower error than the existing state-of-the-art methods.

Simulator education in Japanese dental education.

Advances in information and communication technology (ICT) and digital technology have contributed greatly to the development of educational methods, and new educational styles have been introduced over the past several years. In recent years, it has become possible to provide more realistic simulator training to prepare for patient care through the development of phantom heads with jaws and artificial teeth that reflect actual anatomical structures and composition. While these phantom head simulator resources are becoming more realistic, the development of digital devices utilizing virtual reality has enabled training that is even more realistic than the jaw models, such as scanning the teeth and transfer this scan into a VR model in the simulator which can be used to practice on prior to treating the patient. There was a basic research report on the development of VR haptics simulators in Japan about 20 years ago, but there have been no reports for some time since then, and there are several reports on VR haptics simulators published after 2018. Even now, in terms of utilization in dental education, it is difficult to say that VR haptics simulator has been widely introduced in Japan. The reasons for this may be related to the evaluation system for clinical dental skills in the Japanese dental education system, as well as issues with budgets and in current condition of dental clinical practice in dental schools. It must be that the VR haptics simulator will be utilized in terms of preparatory training for patient care as clinical dental practice participation is enhanced in the future.

Validation of a GPS Sensor Vest for Predicting Ground Reaction Force

This study presents a novel approach to estimate Ground Reaction Forces (GRFs) during athletic exercises using wearable GPS sensor technology and artificial neural networks (ANNs). GRFs have typically required the utilization of heavy force platforms in carefully controlled laboratory conditions in order to fully comprehend the biomechanical stresses that athletes experience during training and competition. In contrast, this study makes use of wearable sensors' usefulness to capture real-time data in realistic training and gaming contexts. With excellent R2 scores and small Root Mean Squared Error (RMSE) values for a variety of trials, the constructed ANN model demonstrates impressive accuracy. These results demonstrate that the ANN can successfully establish the complex link between GPS sensor data and GRFs, allowing accurate predictions of these forces during dynamic motions. These results demonstrate that the ANN can successfully establish the complex link between GPS sensor data and GRFs, allowing accurate predictions of these forces during dynamic motions. This work highlights the potential for practical sports science applications, providing coaches, athletes, and medical teams with invaluable information about enhancing performance and reducing injury risks.