Wearable Data‐Driven Transformer Model for Physical Load Assessment in Badminton Training

Remaining useful life (RUL) prediction is crucial in the field of engineering, which can reduce the frequency of accidents and the maintenance cost of machinery. With the increasing complexity of rotating machinery, the data analysis methods based on deep learning have become the mainstream methods of prediction work. However, most of the current RUL prediction methods only use single-sensor data as input, which cannot effectively use multi-sensor data. In addition, as an advanced deep learning prediction method, temporal convolutional network (TCN) only uses the past time information of vibration data to determine the current health status of bearings, while ignoring the importance of future time information of vibration data. To solve the above problems, a bearing RUL prediction method based on multi-sensor data fusion and bidirectional-temporal attention convolutional network (Bi-TACN) is proposed in this paper. In multi-sensor data fusion, multi-sensor data are combined into multi-channel data, and a channel-weighted attention is designed to emphasize the importance of each sensor data. Compared with traditional multi-sensor data fusion, the proposed fusion method allows deep prediction networks to learn more useful feature information from multi-sensor data. Then, Bi-TACN is developed to predict the RUL of bearings. Bi-TACN is mainly composed of the forward TCN block and the backward TCN block, both of which can learn the past and future time information of multi-sensor data simultaneously. Moreover, a temporal attention mechanism is embedded in Bi-TACN to adaptively calibrate the weights of the two TCN blocks, so as to achieve dynamic feature fusion of past and future time information. RUL prediction experiments are carried out through Xi’an Jiao tong University bearing dataset and PHM 2012 bearing dataset respectively. Compared with the advanced prediction methods, the proposed method can accurately predict the RUL of more types of bearings and has low prediction errors.

Existing deep learning based methods for 3D human pose estimation cannot be deployed on resource-constrained devices. In this paper, we propose a Tiny-HourglassNet for 3D human pose estimation that improves the efficiency of Stacked Hourglass Networks with a guarantee of estimation performance. To be concrete, we develop an efficient Tiny-Hourglass backbone with two different types of ShuffleNet V2 blocks. At the meantime, we present a Simple Estimation Head (SEH) to reduce the resource utility for depth prediction. Furthermore, we develop two feature enhancement modules, namely Feature Enhancement Module (FEM) and Intermediate Enhancement Module (IEM), to improve the feature representation ability of Tiny-HourglassNet without evident increment of resource usage and computational complexity. Experimental results demonstrate that Tiny-HourglassNet achieves equivalent accuracy with a reduction of 80% complexity (FLOPs) in comparison to the baselines.

The non-intrusive load decomposition method helps users understand the current situation of electricity consumption and reduce energy consumption. Traditional methods based on deep learning are difficult to identify low usage appliances, and are prone to model degradation leading to insufficient classification capacity. To solve this problem, this paper proposes a dilated residual aggregation network to achieve non-intrusive load decomposition. First, the original power data is processed by difference to enhance the data expression ability. Secondly, the residual structure and dilated convolution are combined to realize the cross layer transmission of load characteristic information, and capture more long sequence content. Then, the feature enhancement module is proposed to recalibrate the local feature mapping, so as to enhance the learning ability of its own network for subtle features. Compared to traditional network models, the null-residual aggregated convolutional network model has the advantages of strong learning capability for fine load features and good generalisation performance, improving the accuracy of load decomposition. The experimental results on several datasets show that the network model has good generalization performance and improves the recognition accuracy of low usage appliances.

Accurate crop classification is critical for optimizing agricultural resource use and informing production decisions. Deep learning, with its robust feature extraction ability, has become a prevalent technique for remote sensing-based crop classification. However, agricultural landscape complexity poses three key challenges: background noise interference, class confusion from inter-crop spectral similarity, and blurred small-area crop boundaries due to class imbalance. This paper proposes FCR-TransUNet, a TransUNet-based enhanced model integrating three modules: Feature Enhancement Module (FEM) for noise filtering, Class-Attention (CAExperimental results on the Youyi Farm and barley datasets validate the superiority of the proposed model. On the Youyi Farm dataset, FCR-TransUNet achieves an MIoU of 92.2%, representing an improvement of 1.8% over SAM2-UNet and 2.9% over the baseline TransUNet. On the barley dataset, it yields an MIoU of 89.9%. Ablation studies further verify the effectiveness of each designed module. To comprehensively evaluate the classification performance of FCR-TransUNet across the full crop growth cycle, experiments were conducted using remote sensing images from May, July, and August, respectively. The results demonstrate that FCR-TransUNet exhibits strong stability and adaptability at different crop growth stages, providing a reliable solution for precision agriculture and intelligent agricultural production.

Defect detection plays a crucial role in the manufacturing industry, ensuring the quality of industrial products. Despite advancements in this field, current defect detection methods face two primary challenges: (1) extracting visually similar features from the background poses difficult and (2) these methods struggle to identify tiny defects in the target objects. We present a feature enhancement module called the SE-CAR, which aims to handle the identified problems effectively. This module is designed to efficiently capture tiny defects in images, prioritize defect information features, and ultimately enhance the model’s predictive performance and accuracy in defect recognition tasks. In addition, Distance-IoU Non-Maximum Suppression is employed as a substitution for the original Non-Maximum Suppression. This enhances the recognition accuracy of bounding boxes, ensuring that the model maintains high detection accuracy even after complex scenarios. Moreover, the proposed methodology exhibits broad applicability across a wide spectrum of prevalent defect detection paradigms. In empirical experiments, we employ the YOLOv7 architecture as the foundation framework, integrating the proposed methodologies for the purpose of detecting defects in the membrane. The empirical evidence demonstrates a notable improvement in the performance of detecting defects in membrane products using the SE-CAR feature enhancement module and Distance-IoU Non-Maximum Suppression algorithm. In comparison to baseline networks, an increase in mAp by 6.29%, precision by 1.63%, and recall by 8.76%.

Background/Objectives: Thymoma is a rare epithelial neoplasm originating from the thymus gland, and its accurate detection and classification using computed tomography (CT) images remain diagnostically challenging due to subtle morphological similarities with other mediastinal pathologies. This study presents a deep learning (DL)-based model designed to improve diagnostic accuracy for both thymoma detection and subtype classification (benign vs. malignant). Methods: The proposed approach integrates a pre-trained VGG16 network for efficient feature extraction—capitalizing on its capacity to capture hierarchical spatial features—and an MLP-Mixer-based feature enhancement module, which effectively models both local and global feature dependencies without relying on conventional convolutional mechanisms. Additionally, customized preprocessing and post-processing methods are employed to enhance image quality and suppress redundant data. The model’s performance was evaluated on two classification tasks: distinguishing thymoma from healthy cases and discriminating between benign and malignant thymoma. Comparative analysis was conducted against state-of-the-art DL models including ResNet50, ResNet34, SEResNeXt50, InceptionResNetV2, MobileNetV2, VGG16, InceptionV3, and DenseNet121 using metrics such as F1 score, accuracy, recall, and precision. Results: The model proposed in this study obtained its best performance in thymoma vs. healthy classification, with an accuracy of 97.15% and F1 score of 80.99%. In the benign vs. malignant task, it attained an accuracy of 79.20% and an F1 score of 78.51%, outperforming all baseline methods. Conclusions: The integration of VGG16’s robust spatial feature extraction and the MLP-Mixer’s effective feature mixing demonstrates superior and balanced performance, highlighting the model’s potential for clinical decision support in thymoma diagnosis.

The rapid development and progress in deep machine-learning techniques have become a key factor in solving the future challenges of humanity. Vision-based target detection and object classification have been improved due to the development of deep learning algorithms. Data fusion in autonomous driving is a fact and a prerequisite task of data preprocessing from multi-sensors that provide a precise, well-engineered, and complete detection of objects, scene or events. The target of the current study is to develop an in-vehicle information system to prevent or at least mitigate traffic issues related to parking detection and traffic congestion detection. In this study we examined to solve these problems described by (1) extracting region-of-interest in the images (2) vehicle detection based on instance segmentation, and (3) building deep learning model based on the key features obtained from input parking images. We build a deep machine learning algorithm that enables collecting real video-camera feeds from vision sensors and predicting free parking spaces. Image augmentation techniques were performed using edge detection, cropping, refined by rotating, thresholding, resizing, or color augment to predict the region of bounding boxes. A deep convolutional neural network F-MTCNN model is proposed that simultaneously capable for compiling, training, validating and testing on parking video frames through video-camera. The results of proposed model employing on publicly available PK-Lot parking dataset and the optimized model achieved a relatively higher accuracy 97.6% than previous reported methodologies. Moreover, this article presents mathematical and simulation results using state-of-the-art deep learning technologies for smart parking space detection. The results are verified using Python, TensorFlow, OpenCV computer simulation frameworks.

PURPOSE The purpose of this study was to evaluate the validity and reliability of the portable KB1-C Metabolic Analysis system during submaximal exercise intensities. METHODS Thirteen apparently health subjects (age: 21.9 ± 0.4 yrs; height: 175.1 ± 2.6 cm; weight: 75.3 ± 4.3 kg; %BF: 15.0 ± 1.8) were recruited. Each subject exercised on a treadmill and a cycle ergometer. The stages for the treadmill protocol included rest and 2.0, 3.0, 4.0, 5.0 and 6.0 miles per hour (mph). The stages for the cycle ergometer protocol included rest and 50, 75, 100, 125, 150 and 175 Watts. Each subject exercised for five minutes at each exercise intensity. Expired air was collected and simultaneously analyzed for oxygen consumption (Vo2), carbon dioxide production (Vco2), minute ventilation (Ve), fraction of expired oxygen (FEO2) and fraction of expired carbon dioxide (FECO2) by the KB1-C and True Max 2400 computerized metabolic systems. Data collected over the last two minutes were averaged and compared. A two-way ANOVA with repeated measures (instrument × stage) was used to identify differences between the two metabolic systems during each stage of the submaximal exercise test. Pearson's correlation coefficient was used to identify relationships between the two instruments. RESULTS Oxygen consumption, Vco2, and Ve increased significantly (p< 0.05) with increased exercise intensity. During the treadmill test, oxygen consumption and Ve were significantly different between the two systems at rest. VCO2 measured by the KB1-Cwas significantly higher at rest, 2.0, and 4.0 mph compared to the True Max system. On the other hand, FEO2 and FECO2 were not significantly different between systems. During the cycle ergometer test, Vo2, and Ve were not significantly different between instruments at 50, 75 and 100 Watts but KB1-C was significantly higher than the True Max at rest and significantly lower than the True Max at 125, 150 and 175 Watts. VCO2 was significantly different at all exercise intensities between instruments. Fraction of expired oxygen assessed by the KB1-C was significantly higher at 125, 150, and 175 Watts compared to the True Max system. On the other hand, FECO2 was not significantly different between instruments. Finally, correlations for Vo2 values between the two systems were r≥0.85. CONCLUSIONS These results indicate that the KB1-C can be used in the assessment of oxygen consumption, especially at low and moderate exercise intensities. However, at higher exercise intensities, especially if a cycle ergometer is used, KB1-C does not present equally valid results. It is possible that the differences between instruments may be even greater at maximal exercise intensities.

With the development of the power grid Internet of things (IoT), more and more sensor devices access the power grid network, putting a lot of pressure on cloud servers. As an effective solution, edge computing can effectively reduce the possibility of network congestion and improve the delay of sensor perception. However, the storage capacity and coverage area of edge node is limited. In this paper, an edge perception strategy based on data fusion and recurrent neural network is proposed. Considering the features of sensor devices in edge area, we first propose a data preprocessing scheme. In order to fuse congeneric data in a region, we then proposed a data fusion method based on self-adaptive average weighting algorithm. Furthermore, a perceptual prediction method based on recurrent neural network (RNN) is proposed. Simulation results show that our proposed strategy can effectively complete sensor data fusion and perception in edge area, and extend the perceptive scope of the edge node.KeywordsEdge perception strategyPower grid Internet of thingsData fusionRecurrent neural network

Soil moisture (SM) monitoring in farmland at a regional scale is crucial for precision irrigation management and ensuring food security. However, existing methods for SM estimation encounter significant challenges related to accuracy, generalizability, and automation. This study proposes an integrated data fusion method to systematically assess the potential of three automated machine learning (AutoML) frameworks—tree-based pipeline optimization tool (TPOT), AutoGluon, and H2O AutoML—in retrieving SM. To evaluate the impact of input variables on estimation accuracy, six input scenarios were designed: multispectral data (MS), thermal infrared data (TIR), MS combined with TIR, MS with auxiliary data, TIR with auxiliary data, and a comprehensive combination of MS, TIR, and auxiliary data. The research was conducted in a winter wheat cultivation area within the People’s Victory Canal Irrigation Area, focusing on the 0–40 cm soil layer. The results revealed that the scenario incorporating all data types (MS + TIR + auxiliary) achieved the highest retrieval accuracy. Under this scenario, all three AutoML frameworks demonstrated optimal performance. AutoGluon demonstrated superior performance in most scenarios, particularly excelling in the MS + TIR + auxiliary data scenario. It achieved the highest retrieval accuracy with a Pearson correlation coefficient (R) value of 0.822, root mean square error (RMSE) of 0.038 cm3/cm3, and relative root mean square error (RRMSE) of 16.46%. This study underscores the critical role of input data types and fusion strategies in enhancing SM estimation accuracy and highlights the significant advantages of AutoML frameworks for regional-scale SM retrieval. The findings offer a robust technical foundation and theoretical guidance for advancing precision irrigation management and efficient SM monitoring.

Investigating the learning approaches of students in nursing education

Improving the performance of machine learning algorithms to overcome the curse of dimensionality while maintaining interpretability is still a challenging issue for researchers in artificial intelligence. Patch learning (PL), based on the improved adaptive network-based fuzzy inference system (ANFIS) and continuous local optimization for the input domain, is characterized by high accuracy. However, PL can only handle low-dimensional data set regression. Based on the parallel and serial ensembles, two deep patch learning algorithms with embedded adaptive fuzzy systems (DPLFSs) are proposed in this paper. First, using the maximum information coefficient (MIC) and Pearson's correlation coefficients for feature selection, the variables with the least relationship (linear or nonlinear) are excluded. Second, principal component analysis is used to reduce the complexity further of DPLFSs. Meanwhile, fuzzy C-means clustering is used to enhance the interpretability of DPLFSs. Then, an improved PL method is put forward for the training of each sub-fuzzy system in a fashion of bottom-up layer-by-layer, and finally, the structure optimization is performed to significantly improve the interpretability of DPLFSs. Experiments on several benchmark data sets show the advantages of a DPLFS: (1) it can handle medium-scale data sets; (2) it can overcome the curse of dimensionality faced by PL; (3) its precision and generalization are greatly improved; and (4) it can overcome the poor interpretability of deep learning networks. Compared with shallow and deep learning algorithms, DPLFSs have the advantages of interpretability, self-learning, and high precision. DPLFS1 is superior for medium-scale data; DPLFS2 is more efficient and effective for high-dimensional problems, has a faster convergence, and is more interpretable.

Conventional wearable monitoring devices often suffer from insufficient data accuracy and low posture recognition rates, making them inadequate for the demands of professional sports health monitoring. To address these issues, this study proposes a wearable monitoring system for running gait analysis based on the Diffusion Transformer (DiT). The system aims to achieve high-precision running posture recognition and real-time motion monitoring through multi-sensor data fusion and advanced deep learning architecture. First, a wearable system was developed using a nine-axis Micro-Electro-Mechanical System (MEMS) inertial sensor and an UltraWide Band (UWB) positioning module. Data quality was enhanced through sensor calibration, noise compensation, and an adaptive filtering algorithm. Then, a DiT-LSTM running posture recognition model was constructed by integrating the DiT with a Long Short-Term Memory (LSTM) network to perform posture recognition within the wearable system. Experimental results show that on the Human3.6M dataset, the DiT-LSTM model achieved an accuracy of 97.54%, a precision of 97.61%, a recall of 97.73%, an F1-score of 97.58%, and an Area Under the Curve (AUC) of 98.61%. On the HumanEva dataset, the model attained an accuracy of 96.39%, a precision of 96.47%, a recall of 96.8%, an F1-score of 96.92%, and an AUC of 97.9%, all outperforming other algorithms. The complexity assessment showed that DiT-LSTM reached 14.7 GFLOPs on the Human3.6M dataset, with a per-epoch training time of 62.3 seconds, while its per-sample inference latency was only 9.4 ms, meeting real-time monitoring requirements. In the UWB-based position drift correction experiment, Sample 1 achieved mean errors of 0.637 m, 0.581 m, and 0.349 m on the X/Y/Z axes, with corresponding RMSE values of 0.041 m, 0.023 m, and 0.025 m, demonstrating high positioning accuracy and stability. By combining multimodal sensors with the DiT-LSTM model, the study offers reliable technical support for running gait analysis, injury prevention, and personalized training guidance.

Knee segmentation and landmark localization from 3D MRI are two significant tasks for diagnosis and treatment of knee diseases. With the development of deep learning, Convolutional Neural Network (CNN) based methods have become the mainstream. However, the existing CNN methods are mostly single-task methods. Due to the complex structure of bone, cartilage and ligament in the knee, it is challenging to complete the segmentation or landmark localization alone. And establishing independent models for all tasks will bring difficulties for surgeon's clinical using. In this paper, a Spatial Dependence Multi-task Transformer (SDMT) network is proposed for 3D knee MRI segmentation and landmark localization. We use a shared encoder for feature extraction, then SDMT utilizes the spatial dependence of segmentation results and landmark position to mutually promote the two tasks. Specifically, SDMT adds spatial encoding to the features, and a task hybrided multi-head attention mechanism is designed, in which the attention heads are divided into the inter-task attention head and the intra-task attention head. The two attention head deal with the spatial dependence between two tasks and correlation within the single task, respectively. Finally, we design a dynamic weight multi-task loss function to balance the training process of two task. The proposed method is validated on our 3D knee MRI multi-task datasets. Dice can reach 83.91% in the segmentation task, and MRE can reach 2.12 mm in the landmark localization task, it is competitive and superior over other state-of-the-art single-task methods.

GeoAI in terrain analysis: Enabling multi-source deep learning and data fusion for natural feature detection