Improved Chinese Giant Salamander Parental Care Behavior Detection Based on YOLOv8.

Optimizing the breeding techniques and increasing the hatching rate of Andrias davidianus offspring necessitates a thorough understanding of its parental care behaviors. However, A. davidianus' nocturnal and cave-dwelling tendencies pose significant challenges for direct observation. To address this problem, this study constructed a dataset for the parental care behavior of A. davidianus, applied the target detection method to this behavior for the first time, and proposed a detection model for A. davidianus' parental care behavior based on the YOLOv8s algorithm. Firstly, a multi-scale feature fusion convolution (MSConv) is proposed and combined with a C2f module, which significantly enhances the feature extraction capability of the model. Secondly, the large separable kernel attention is introduced into the spatial pyramid pooling fast (SPPF) layer to effectively reduce the interference factors in the complex environment. Thirdly, to address the problem of low quality of captured images, Wise-IoU (WIoU) is used to replace CIoU in the original YOLOv8 to optimize the loss function and improve the model's robustness. The experimental results show that the model achieves 85.7% in the mAP50-95, surpassing the YOLOv8s model by 2.1%. Compared with other mainstream models, the overall performance of our model is much better and can effectively detect the parental care behavior of A. davidianus. Our research method not only offers a reference for the behavior recognition of A. davidianus and other amphibians but also provides a new strategy for the smart breeding of A. davidianus.

Coreference Resolution Based on High-Dimensional Multi-Scale Information.

Coreference resolution is a key task in Natural Language Processing. It is difficult to evaluate the similarity of long-span texts, which makes text-level encoding somewhat challenging. This paper first compares the impact of commonly used methods to improve the global information collection ability of the model on the BERT encoding performance. Based on this, a multi-scale context information module is designed to improve the applicability of the BERT encoding model under different text spans. In addition, improving linear separability through dimension expansion. Finally, cross-entropy loss is used as the loss function. After adding BERT and span BERT to the module designed in this article, F1 increased by 0.5% and 0.2%, respectively.

Plane-wave medical image reconstruction based on dynamic Criss-Cross attention and multi-scale convolution.

BACKGROUND: Plane-wave imaging is widely employed in medical imaging due to its ultra-fast imaging speed. However, the image quality is compromised. Existing techniques to enhance image quality tend to sacrifice the imaging frame rate. OBJECTIVE: The study aims to reconstruct high-quality plane-wave images while maintaining the imaging frame rate. METHODS: The proposed method utilizes a U-Net-based generator incorporating a multi-scale convolution module in the encoder to extract information at different levels. Additionally, a Dynamic Criss-Cross Attention (DCCA) mechanism is proposed in the decoder of the U-Net-based generator to extract both local and global features of plane-wave images while avoiding interference caused by irrelevant regions. RESULTS: In the reconstruction of point targets, the experimental images achieved a reduction in Full Width at Half Maximum (FWHM) of 0.0499 mm, compared to the Coherent Plane-Wave Compounding (CPWC) method using 75-beam plane waves. For the reconstruction of cyst targets, the simulated image achieved a 3.78% improvement in Contrast Ratio (CR) compared to CPWC. CONCLUSIONS: The proposed model effectively addresses the issue of unclear lesion sites in plane-wave images.

Segmenting Ischemic Penumbra and Infarct Core Simultaneously on Non-Contrast CT of Patients with Acute Ischemic Stroke Using Novel Convolutional Neural Network.

Differentiating between a salvageable Ischemic Penumbra (IP) and an irreversibly damaged Infarct Core (IC) is important for therapy decision making for acute ischemic stroke (AIS) patients. Existing methods rely on Computed Tomography Perfusion (CTP) or Diffusion-Weighted Imaging-Fluid Attenuated Inversion Recovery (DWI-FLAIR). We designed a novel Convolutional Neural Network named I2PC-Net, which relies solely on Non-Contrast Computed Tomography (NCCT) for the automatic and simultaneous segmentation of the IP and IC. In the encoder, Multi-Scale Convolution (MSC) blocks were proposed to capture effective features of ischemic lesions, and in the deep levels of the encoder, Symmetry Enhancement (SE) blocks were also designed to enhance anatomical symmetries. In the attention-based decoder, hierarchical deep supervision was introduced to address the challenge of differentiating between the IP and IC. We collected 197 NCCT scans from AIS patients to evaluate the proposed method. On the test set, I2PC-Net achieved Dice Similarity Scores of 42.76 ± 21.84%, 33.54 ± 24.13% and 65.67 ± 12.30% and lesion volume correlation coefficients of 0.95 (p < 0.001), 0.61 (p < 0.001) and 0.93 (p < 0.001) for the IP, IC and IP + IC, respectively. The results indicated that NCCT could potentially be used as a surrogate technique of CTP for the quantitative evaluation of the IP and IC.

A Sequential End-to-End Neonatal Sleep Staging Model with Squeeze and Excitation Blocks and Sequential Multi-Scale Convolution Neural Networks.

Automatic sleep staging offers a quick and objective assessment for quantitatively interpreting sleep stages in neonates. However, most of the existing studies either do not encompass any temporal information, or simply apply neural networks to exploit temporal information at the expense of high computational overhead and modeling ambiguity. This limits the application of these methods to multiple scenarios. In this paper, a sequential end-to-end sleep staging model, SeqEESleepNet, which is competent for parallelly processing sequential epochs and has a fast training rate to adapt to different scenarios, is proposed. SeqEESleepNet consists of a sequence epoch generation (SEG) module, a sequential multi-scale convolution neural network (SMSCNN) and squeeze and excitation (SE) blocks. The SEG module expands independent epochs into sequential signals, enabling the model to learn the temporal information between sleep stages. SMSCNN is a multi-scale convolution neural network that can extract both multi-scale features and temporal information from the signal. Subsequently, the followed SE block can reassign the weights of features through mapping and pooling. Experimental results exhibit that in a clinical dataset, the proposed method outperforms the state-of-the-art approaches, achieving an overall accuracy, F1-score, and Kappa coefficient of 71.8%, 71.8%, and 0.684 on a three-class classification task with a single channel EEG signal. Based on our overall results, we believe the proposed method could pave the way for convenient multi-scenario neonatal sleep staging methods.

4mCPred-GSIMP: Predicting DNA N4-methylcytosine sites in the mouse genome with multi-Scale adaptive features extraction and fusion.

The epigenetic modification of DNA N4-methylcytosine (4mC) is vital for controlling DNA replication and expression. It is crucial to pinpoint 4mC's location to comprehend its role in physiological and pathological processes. However, accurate 4mC detection is difficult to achieve due to technical constraints. In this paper, we propose a deep learning-based approach 4mCPred-GSIMP for predicting 4mC sites in the mouse genome. The approach encodes DNA sequences using four feature encoding methods and combines multi-scale convolution and improved selective kernel convolution to adaptively extract and fuse features from different scales, thereby improving feature representation and optimization effect. In addition, we also use convolutional residual connections, global response normalization and pointwise convolution techniques to optimize the model. On the independent test dataset, 4mCPred-GSIMP shows high sensitivity, specificity, accuracy, Matthews correlation coefficient and area under the curve, which are 0.7812, 0.9312, 0.8562, 0.7207 and 0.9233, respectively. Various experiments demonstrate that 4mCPred-GSIMP outperforms existing prediction tools.

Rapid diagnosis of rheumatoid arthritis and ankylosing spondylitis based on Fourier transform infrared spectroscopy and deep learning.

OBJECTIVE: Rheumatoid arthritis and Ankylosing spondylitis are two common autoimmune inflammatory rheumatic diseases that negatively affect activities of daily living and can lead to structural and functional disability, reduced quality of life. Here, this study utilized Fourier transform infrared (FTIR) spectroscopy on dried serum samples and achieved early diagnosis of rheumatoid arthritis and ankylosing spondylitis based on deep learning models. METHOD: A total of 243 dried serum samples were collected in this study, including 81 samples each from ankylosing spondylitis, rheumatoid arthritis, and healthy controls. Three multi-scale convolutional modules with different specifications were designed based on the multi-scale convolutional neural network (MSCNN) to effectively fuse the local features to enhance the generalization ability of the model. The FTIR was then combined with the MSCNN model to achieve a non-invasive, fast, and accurate diagnosis of ankylosing spondylitis, rheumatoid arthritis, and healthy controls. RESULTS: Spectral analysis shows that the curves and waveforms of the three spectral graphs are similar. The main differences are distributed in the spectral regions of 3300-3250 cm-1, 3000-2800 cm-1, 1750-1500 cm-1, and 1500-1300 cm-1, which represent: Amides, fatty acids, cholesterol, proteins with a carboxyl group, amide II, free amino acids, and polysaccharides. Four classification models, namely artificial neural network (ANN), convolutional neural network (CNN), improved AlexNet model, and multi-scale convolutional neural network (MSCNN) were established. Through comparison, it was found that the diagnostic AUC value of the MSCNN model was 0.99, and the accuracy rate was as high as 0.93, which was much higher than the other three models. CONCLUSION: The study demonstrated the superiority of MSCNN in distinguishing ankylosing spondylitis from rheumatoid arthritis and healthy controls. FTIR may become a rapid, sensitive, and non-invasive means of diagnosing rheumatism.

A Framework for Enhancing Stock Investment Performance by Predicting Important Trading Points with Return-Adaptive Piecewise Linear Representation and Batch Attention Multi-Scale Convolutional Recurrent Neural Network.

Efficient stock status analysis and forecasting are important for stock market participants to be able to improve returns and reduce associated risks. However, stock market data are replete with noise and randomness, rendering the task of attaining precise price predictions arduous. Moreover, the lagging phenomenon of price prediction makes it hard for the corresponding trading strategy to capture the turning points, resulting in lower investment returns. To address this issue, we propose a framework for Important Trading Point (ITP) prediction based on Return-Adaptive Piecewise Linear Representation (RA-PLR) and a Batch Attention Multi-Scale Convolution Recurrent Neural Network (Batch-MCRNN) with the starting point of improving stock investment returns. Firstly, a novel RA-PLR method is adopted to detect historical ITPs in the stock market. Then, we apply the Batch-MCRNN model to integrate the information of the data across space, time, and sample dimensions for predicting future ITPs. Finally, we design a trading strategy that combines the Relative Strength Index (RSI) and the Double Check (DC) method to match ITP predictions. We conducted a comprehensive and systematic comparison with several state-of-the-art benchmark models on real-world datasets regarding prediction accuracy, risk, return, and other indicators. Our proposed method significantly outperformed the comparative methods on all indicators and has significant reference value for stock investment.

A deep learning approach for inpatient length of stay and mortality prediction.

PURPOSE: Accurate prediction of the Length of Stay (LoS) and mortality in the Intensive Care Unit (ICU) is crucial for effective hospital management, and it can assist clinicians for real-time demand capacity (RTDC) administration, thereby improving healthcare quality and service levels. METHODS: This paper proposes a novel one-dimensional (1D) multi-scale convolutional neural network architecture, namely 1D-MSNet, to predict inpatients' LoS and mortality in ICU. First, a 1D multi-scale convolution framework is proposed to enlarge the convolutional receptive fields and enhance the richness of the convolutional features. Following the convolutional layers, an atrous causal spatial pyramid pooling (SPP) module is incorporated into the networks to extract high-level features. The optimized Focal Loss (FL) function is combined with the synthetic minority over-sampling technique (SMOTE) to mitigate the imbalanced-class issue. RESULTS: On the MIMIC-IV v1.0 benchmark dataset, the proposed approach achieves the optimum R-Square and RMSE values of 0.57 and 3.61 for the LoS prediction, and the highest test accuracy of 97.73% for the mortality prediction. CONCLUSION: The proposed approach presents a superior performance in comparison with other state-of-the-art, and it can effectively perform the LoS and mortality prediction tasks.

EEG Emotion Recognition by Fusion of Multi-Scale Features.

Electroencephalogram (EEG) signals exhibit low amplitude, complex background noise, randomness, and significant inter-individual differences, which pose challenges in extracting sufficient features and can lead to information loss during the mapping process from low-dimensional feature matrices to high-dimensional ones in emotion recognition algorithms. In this paper, we propose a Multi-scale Deformable Convolutional Interacting Attention Network based on Residual Network (MDCNAResnet) for EEG-based emotion recognition. Firstly, we extract differential entropy features from different channels of EEG signals and construct a three-dimensional feature matrix based on the relative positions of electrode channels. Secondly, we utilize deformable convolution (DCN) to extract high-level abstract features by replacing standard convolution with deformable convolution, enhancing the modeling capability of the convolutional neural network for irregular targets. Then, we develop the Bottom-Up Feature Pyramid Network (BU-FPN) to extract multi-scale data features, enabling complementary information from different levels in the neural network, while optimizing the feature extraction process using Efficient Channel Attention (ECANet). Finally, we combine the MDCNAResnet with a Bidirectional Gated Recurrent Unit (BiGRU) to further capture the contextual semantic information of EEG signals. Experimental results on the DEAP dataset demonstrate the effectiveness of our approach, achieving accuracies of 98.63% and 98.89% for Valence and Arousal dimensions, respectively.

[Novel Pulmonary Nodule Position Detection Method Based on Multiscale Convolution].

OBJECTIVE: In order to improve the accuracy of the current pulmonary nodule location detection method based on CT images, reduce the problem of missed detection or false detection, and effectively assist imaging doctors in the diagnosis of pulmonary nodules. METHODS: Propose a novel method for detecting the location of pulmonary nodules based on multiscale convolution. First, image preprocessing methods are used to eliminate the noise and artifacts in lung CT images. Second, multiple adjacent single-frame CT images are selected to be concatenate into multi-frame images, and the feature extraction is carried out through the artificial neural network model U-Net improved by multi-scale convolution to enhanced feature extraction capability for pulmonary nodules of different sizes and shapes, so as to improve the accuracy of feature extraction of pulmonary nodules. Finally, using point detection to improve the loss function of U-Net training process, the accuracy of pulmonary nodule location detection is improved. RESULTS: The accuracy of detecting pulmonary nodules equal or larger than 3 mm and smaller than 3 mm are 98.02% and 96.94% respectively. CONCLUSIONS: This method can effectively improve the detection accuracy of pulmonary nodules on CT image sequence, and can better meet the diagnostic needs of pulmonary nodules.

A Joint De-Rain and De-Mist Network Based on the Atmospheric Scattering Model.

Rain can have a detrimental effect on optical components, leading to the appearance of streaks and halos in images captured during rainy conditions. These visual distortions caused by rain and mist contribute significant noise information that can compromise image quality. In this paper, we propose a novel approach for simultaneously removing both streaks and halos from the image to produce clear results. First, based on the principle of atmospheric scattering, a rain and mist model is proposed to initially remove the streaks and halos from the image by reconstructing the image. The Deep Memory Block (DMB) selectively extracts the rain layer transfer spectrum and the mist layer transfer spectrum from the rainy image to separate these layers. Then, the Multi-scale Convolution Block (MCB) receives the reconstructed images and extracts both structural and detailed features to enhance the overall accuracy and robustness of the model. Ultimately, extensive results demonstrate that our proposed model JDDN (Joint De-rain and De-mist Network) outperforms current state-of-the-art deep learning methods on synthetic datasets as well as real-world datasets, with an average improvement of 0.29 dB on the heavy-rainy-image dataset.

MSATNet: multi-scale adaptive transformer network for motor imagery classification.

Motor imagery brain-computer interface (MI-BCI) can parse user motor imagery to achieve wheelchair control or motion control for smart prostheses. However, problems of poor feature extraction and low cross-subject performance exist in the model for motor imagery classification tasks. To address these problems, we propose a multi-scale adaptive transformer network (MSATNet) for motor imagery classification. Therein, we design a multi-scale feature extraction (MSFE) module to extract multi-band highly-discriminative features. Through the adaptive temporal transformer (ATT) module, the temporal decoder and multi-head attention unit are used to adaptively extract temporal dependencies. Efficient transfer learning is achieved by fine-tuning target subject data through the subject adapter (SA) module. Within-subject and cross-subject experiments are performed to evaluate the classification performance of the model on the BCI Competition IV 2a and 2b datasets. The MSATNet outperforms benchmark models in classification performance, reaching 81.75 and 89.34% accuracies for the within-subject experiments and 81.33 and 86.23% accuracies for the cross-subject experiments. The experimental results demonstrate that the proposed method can help build a more accurate MI-BCI system.

An EIT image reconstruction method based on DenseNet with multi-scale convolution.

Electrical impedance tomography (EIT) is an imaging technique that non-invasively acquires the electrical conductivity distribution within a field. The ill-posed and nonlinear nature of the image reconstruction process results in lower quality of the obtained images. To solve this problem, an EIT image reconstruction method based on DenseNet with multi-scale convolution named MS-DenseNet is proposed. In the proposed method, three different multi-scale convolutional dense blocks are incorporated to replace the conventional dense blocks; they are placed in parallel to improve the generalization ability of the network. The connection layer between dense blocks adopts a hybrid pooling structure, which reduces the loss of information in the traditional pooling process. A learning rate setting achieves reduction in two stages and optimizes the fitting ability of the network. The input of the constructed network is the boundary voltage data, and the output is the conductivity distribution of the imaging area. The network was trained and tested on a simulated dataset, and it was further tested using actual measurement data. The images reconstructed via this method were evaluated by employing root mean square error, structural similarity index measure, mean absolute error and image correlation coefficient in comparison with conventional DenseNet and Gauss-Newton. The results show that the method improves the artifact and edge blur problems, achieves higher values on the image metrics and improves the EIT image quality.

[Automated detection of sleep-arousal using multi-scale convolution and self-attention mechanism].

In clinical, manually scoring by technician is the major method for sleep arousal detection. This method is time-consuming and subjective. This study aimed to achieve an end-to-end sleep-arousal events detection by constructing a convolutional neural network based on multi-scale convolutional layers and self-attention mechanism, and using 1 min single-channel electroencephalogram (EEG) signals as its input. Compared with the performance of the baseline model, the results of the proposed method showed that the mean area under the precision-recall curve and area under the receiver operating characteristic were both improved by 7%. Furthermore, we also compared the effects of single modality and multi-modality on the performance of the proposed model. The results revealed the power of single-channel EEG signals in automatic sleep arousal detection. However, the simple combination of multi-modality signals may be counterproductive to the improvement of model performance. Finally, we also explored the scalability of the proposed model and transferred the model into the automated sleep staging task in the same dataset. The average accuracy of 73% also suggested the power of the proposed method in task transferring. This study provides a potential solution for the development of portable sleep monitoring and paves a way for the automatic sleep data analysis using the transfer learning method.

MS-Net: Sleep apnea detection in PPG using multi-scale block and shadow module one-dimensional convolutional neural network.

Sleep Apnea (SA) is a respiratory disorder that affects sleep. However, the SA detection method based on polysomnography is complex and not suitable for home use. The detection approach using Photoplethysmography is low cost and convenient, which can be used to widely detect SA. This study proposed a method combining a multi-scale one-dimensional convolutional neural network and a shadow one-dimensional convolutional neural network based on dual-channel input. The time-series feature information of different segments were extracted from multi-scale temporal structure. Moreover, shadow module was adopted to make full use of the redundant information generated after multi-scale convolution operation, which improved the accuracy and ensured the portability of the model. At the same time, we introduced balanced bootstrapping and class weight, which effectively alleviated the problem of unbalanced classes. Our method achieved the result of 82.0% average accuracy, 74.4% average sensitivity and 85.1% average specificity for per-segment SA detection, and reached 93.6% average accuracy for per-recording SA detection after 5-fold cross validation. Experimental results show that this method has good robustness. It can be regarded as an effective aid in SA detection in household use.

Improving the prediction of DNA-protein binding by integrating multi-scale dense convolutional network with fault-tolerant coding.

Accurate prediction of DNA-protein binding (DPB) is of great biological significance for studying the regulatory mechanism of gene expression. In recent years, with the rapid development of deep learning techniques, advanced deep neural networks have been introduced into the field and shown to significantly improve the prediction performance of DPB. However, these methods are primarily based on the DNA sequences measured by the ChIP-seq technology, failing to consider the possible partial variations of the motif sequences and errors of the sequencing technology itself. To address this, we propose a novel computational method, termed MSDenseNet, which combines a new fault-tolerant coding (FTC) scheme with the dense connectional deep neural networks. Three important factors can be attributed to the success of MSDenseNet: First, MSDenseNet utilizes a powerful feature representation approach, which transforms the raw DNA sequence into fusion coding using the fault-tolerant feature sequence; Second, in terms of network structure, MSDenseNet uses a multi-scale convolution within the dense layer and the multi-scale convolution preceding the dense block. This is shown to be able to significantly improve the network performance and accelerate the network convergence speed, and third, building upon the advanced deep neural network, MSDenseNet is capable of effectively mining the hidden complex relationship between the internal attributes of fusion sequence features to enhance the prediction of DPB. Benchmarking experiments on 690 ChIP-seq datasets show that MSDenseNet achieves an average AUC of 0.933 and outperforms the state-of-the-art method. The source code of MSDenseNet is available at https://github.com/csbio-njust-edu/msdensenet. The results show that MSDenseNet can effectively predict DPB. We anticipate that MSDenseNet will be exploited as a powerful tool to facilitate a more exhaustive understanding of DNA-binding proteins and help toward their functional characterization.

DFSNE-Net: Deviant feature sensitive noise estimate network for low-dose CT denoising.

BACKGROUND: Computed tomography (CT) has radiation problems, and high-quality CT scanning is usually accompanied by high doses of radiation that can be harmful to humans, so low-dose CT denoising has received extensive academic attention. METHOD: In this paper, firstly, the concept of deviant features is proposed which lead to the hypothesis of correlation between noise and deviant features. Secondly, in order to estimate the noise in CT, a new method of deviant feature perception and downsampling is proposed. Specifically, the deviant feature perception module based on the multi-scale convolutional cooperative (MSC-DFPM) and Filtering module based on the self-information space attention(SISA-FM) are proposed, and construct the deviant feature sensitive noise estimate network (DFSNE-Net), then a balanced loss function and training strategy adapted to the DFSNE-Net are proposed. Finally, noise distribution normalization based on skewness and kurtosis(SK-NDN) and low credible noise suppression based on confidence interval(CI-LCNS) as the noise optimization methods are proposed to optimize the estimated noise of DFSNE-Net, which is applied to the denoising task of CT. CONCLUSIONS: Experiments and results demonstrate that our proposed denoising method in this paper obtains a better denoising effect than the current state-of-the-art denoising methods in different evaluation indicator, which proves the hypothesis that noise is strongly correlated with deviant features. It is also proved that the denoising method proposed in this paper can achieve the denoising task for different doses of CT.

Automatic detection of multiple types of pneumonia: Open dataset and a multi-scale attention network.

The quick and precise identification of COVID-19 pneumonia, non-COVID-19 viral pneumonia, bacterial pneumonia, mycoplasma pneumonia, and normal lung on chest CT images play a crucial role in timely quarantine and medical treatment. However, manual identification is subject to potential misinterpretations and time-consumption issues owing the visual similarities of pneumonia lesions. In this study, we propose a novel multi-scale attention network (MSANet) based on a bag of advanced deep learning techniques for the automatic classification of COVID-19 and multiple types of pneumonia. The proposed method can automatically pay attention to discriminative information and multi-scale features of pneumonia lesions for better classification. The experimental results show that the proposed MSANet can achieve an overall precision of 97.31%, recall of 96.18%, F1-score of 96.71%, accuracy of 97.46%, and macro-average area under the receiver operating characteristic curve (AUC) of 0.9981 to distinguish between multiple classes of pneumonia. These promising results indicate that the proposed method can significantly assist physicians and radiologists in medical diagnosis. The dataset is publicly available at https://doi.org/10.17632/rf8x3wp6ss.1.

Time series classification with Echo Memory Networks.

Echo state networks (ESNs) are randomly connected recurrent neural networks (RNNs) that can be used as a temporal kernel for modeling time series data, and have been successfully applied on time series prediction tasks. Recently, ESNs have been applied to time series classification (TSC) tasks. However, previous ESN-based classifiers involve either training the model by predicting the next item of a sequence, or predicting the class label at each time step. The former is essentially a predictive model adapted from time series prediction work, rather than a model designed specifically for the classification task. The latter approach only considers local patterns at each time step and then averages over the classifications. Hence, rather than selecting the most discriminating sections of the time series, this approach will incorporate non-discriminative information into the classification, reducing accuracy. In this paper, we propose a novel end-to-end framework called the Echo Memory Network (EMN) in which the time series dynamics and multi-scale discriminative features are efficiently learned from an unrolled echo memory using multi-scale convolution and max-over-time pooling. First, the time series data are projected into the high dimensional nonlinear space of the reservoir and the echo states are collected into the echo memory matrix, followed by a single multi-scale convolutional layer to extract multi-scale features from the echo memory matrix. Max-over-time pooling is used to maintain temporal invariance and select the most important local patterns. Finally, a fully-connected hidden layer feeds into a softmax layer for classification. This architecture is applied to both time series classification and human action recognition datasets. For the human action recognition datasets, we divide the action data into five different components of the human body, and propose two spatial information fusion strategies to integrate the spatial information over them. With one training-free recurrent layer and only one layer of convolution, the EMN is a very efficient end-to-end model, and ranks first in overall classification ability on 55 TSC benchmark datasets and four 3D skeleton-based human action recognition tasks.