Search

Feature representation is a crucial issue in multimodal image registration. The handcrafted features extracted by traditional methods are highly sensitive to nonlinear radiation differences, while supervised learning methods are limited by deficient labeled samples in the remote sensing field. Therefore, this article proposes a consistency feature representation learning method for multimodal image registration, which involves mapping data into a common feature space to realize the accurate alignment of remote sensing images. First, a contrastive network with a spatial attention mechanism is driven to enhance the capability to highlight high-level features of images. Second, a positive sample augmentation strategy is implemented with contrastive loss, which helps the model learn the inherent features better, and imposes constraints on the sample similarity to optimize the feature projection. Finally, a multimodal image registration framework is proposed to enhance the stability of feature matching. The proposed framework achieves accurate feature extraction and consistency feature description for multimodal images, ensuring robustness against nonlinear radiometric differences. Experimental results demonstrate that the proposed method obtains more reliable registration results on the SEN1-2 dataset. Furthermore, the proposed algorithm achieves superior performance on data from other modalities, indicating strong generalization ability.

As a kind of natural disasters, geological hazards pose serious threats to the ecological security and socio-economic development of human society. Evaluating the sensitivity of regional geological hazards is essential for making disaster-preventing/mitigating strategies and land use plans. Based on multisource remote sensing data, this study evaluates the geological hazard sensitivity in Liaoning Province by using the method of spatial principal component analysis and spatial autocorrelation finding that such sensitivity is generally moderate, with an average value of 5.25, and is mainly concentrated in three levels: Mild sensitivity, moderate sensitivity and insensitivity, accounting for more than 67% of the total area. Liaodong Mountain Area and Liaoxi Hilly Area are found with the highest sensitivity of geological hazards, showcasing obvious spatial differences and aggregations, mainly because these areas have large terrain fluctuations, poor stability of rock and soil bodies, and strong erosion by precipitation. The main factors affecting the sensitivity of geological hazards include land use mode, elevation, slope, and lithology. Among them, the land use mode has the highest contribution rate to the first principal component, reaching 0.86. This study provides a theoretical reference for the prevention and control over geological hazards and the protection of ecological security in Liaoning Province.

Objective: Blood circulation is an important indicator of wound healing. In this study, a tissue oxygen saturation detecting (TOSD) system that is based on multispectral imaging (MSI) is proposed to quantify the degree of tissue oxygen saturation (StO2) in cutaneous tissues. Methods: A wound segmentation algorithm is used to segment automatically wound and skin areas, eliminating the need for manual labeling and applying adaptive tissue optics. Animal experiments were conducted on six mice in which they were observed seven times, once every two days. The TOSD system illuminated cutaneous tissues with two wavelengths of light - red ( $\mathrm {\lambda } = 660$ nm) and near-infrared ( $\mathrm {\lambda } = 880$ nm), and StO2 levels were calculated using images that were captured using a monochrome camera. The wound segmentation algorithm using ResNet34-based U-Net was integrated with computer vision techniques to improve its performance. Results: Animal experiments revealed that the wound segmentation algorithm achieved a Dice score of 93.49%. The StO2 levels that were determined using the TOSD system varied significantly among the phases of wound healing. Changes in StO2 levels were detected before laser speckle contrast imaging (LSCI) detected changes in blood flux. Moreover, statistical features that were extracted from the TOSD system and LSCI were utilized in principal component analysis (PCA) to visualize different wound healing phases. The average silhouette coefficients of the TOSD system with segmentation (ResNet34-based U-Net) and LSCI were 0.2890 and 0.0194, respectively. Conclusion: By detecting the StO2 levels of cutaneous tissues using the TOSD system with segmentation, the phases of wound healing were accurately distinguished. This method can support medical personnel in conducting precise wound assessments. Clinical and Translational Impact Statement—This study supports efforts in monitoring StO2 levels, wound segmentation, and wound healing phase classification to improve the efficiency and accuracy of preclinical research in the field.

Due to its ability to reconfigure the optical channel, intelligent reflecting surface (IRS) has revealed its great potential for enhancing the performance of visible light communication (VLC) systems. In this paper, we apply a mirror array-based IRS in typical multiple-input multiple-output based VLC (MIMO-VLC) systems and optimize the IRS configuration to minimize the system bit error rate (BER). By utilizing the tight analytical BER upper bound, a BER minimization problem is formulated subject to the IRS configuration constraints. To solve this problem, an optimization algorithm is proposed which decomposes the original problem into three sub-problems, and the complexity of the algorithm is analyzed. The superiority of the obtained optimal IRS configuration scheme over the baseline scheme in terms of BER performance has been successfully verified through simulations in an indoor IRS-aided $4\times 4$ MIMO-VLC system applying different MIMO techniques.

Neonatal sleep staging is crucial for understanding infant brain development and assessing neurological health. This study explores the optimal electrode configuration to reduce technical complexities and potential risks of causing skin irritation to neonates during data collection. A Multi-Branch Convolutional Neural Network (CNN) is used to categorize neonatal sleep states based on single-channel Electroencephalography (EEG) data. The proposed model was trained and tested on 16803 30-second segments from 64 infants, all of whom were at post-menstrual age between 36 and 43 weeks at the Children’s hospital of Fudan University. A total of 74 extracted time and frequency domain linear and non-linear features are applied to improve the performance of a Multi-Branch CNN-based classification model. Additionally, using principal component analysis (PCA), feature selection and feature importance are also applied to identify the most important features. Notably, the F3 channel outperforms other single-channels and has accuracy and kappa values 74.27±0.80% and 0.61, respectively. Furthermore, a combination of four left-side electrodes yields slightly better classification accuracy (75.36±0.57%) compared to the four right-side electrodes (74.76±1.10%), with corresponding kappa values of 0.63 and 0.62, respectively. In addition to providing insights into optimal electrode configuration using single-channel and multi-channel EEG data, the results highlight the critical role played by specific EEG channels in sleep stage classification. This research has the potential to enhance neonate care and monitor sleep more effectively, enabling early detection of sleep-related abnormalities such as sleep disorders. Furthermore, this research effectively captures information from a single-channel, reducing computing load while maintaining commendable performance. Additionally, integrating time and frequency domain linear and non-linear features into neonatal sleep staging can enhance accuracy and provide a deeper insight into the complex dynamics and irregularities of newborn’s sleep patterns.

Fusing of terrestrial laser scanning (TLS) and airborne laser scanning (ALS) point cloud data has been recognized as an effective approach in forest studies. In this regard, co-registration of point clouds is considered one of the crucial steps in the integration process. Co-registering point clouds in forest environments faces various challenges, including unstable features, extensive occlusions, different viewpoints, and differences in point cloud densities. To address these intricate challenges, this study introduces an automated and robust method for co-registering TLS and ALS point clouds based on the correspondence of individual tree locations in forest environments. Initially, the positions of individual trees in both TLS and ALS data are extracted. Then, a filtering approach is applied to eliminate positions with low potential for corresponding matches in the TLS and ALS dataset. Since larger trees in the TLS data have a higher potential for corresponding matches in the ALS data, an iterative process is applied to identify correspondences between trees in both datasets. After estimating transformation parameters, the co-registration process is executed. The proposed method is applied on six datasets with varying forest complexities. The results demonstrate a high success rate up to 100% if the starting position of the TLS plots are located within ∼4 hectares (∼2000 trees). Additionally, the potential of the proposed method for co-registering TLS data with ALS data across different search areas and varying number of trees is evaluated in detail. The outcomes indicate that successful co-registration of TLS plot with 50 m diameter to ALS data is successful in the best case within a search radius of approximately 113 hectares (∼60,000 tree locations) and in the worst case for around 20 hectares (∼10,000 tree locations) depending on the forest complexity.

Accurate and timely flood forecasting, facilitated by remote sensing technology, is crucial to mitigate the damage and loss of life caused by floods. However, despite years of research, accurate flood prediction still faces numerous challenges, including complex spatiotemporal features and varied flood patterns influenced by multiple variables. Moreover, long-term flood forecasting is always tricky due to the constantly changing conditions of the surrounding environment. In this study, we propose a heterogeneous dynamic temporal graph convolutional network (HD-TGCN) for flood forecasting. Specifically, we designed a dynamic temporal graph convolution module (D-TGCM) to generate a dynamic adjacency matrix by incorporating a multihead self-attention mechanism, enabling our model to capture the dynamic spatiotemporal features of flood data by utilizing temporal graph convolution operations on the dynamic matrix. Furthermore, to reflect the impact of multiple meteorological and hydrological features on the heterogeneity of flood data, we propose a novel approach that utilizes multiple parallel D-TGCMs for processing heterogeneous graph data and implements a fusion mechanism to capture varied flood patterns influenced by multiple variables. Experiments conducted on a real dataset in Wuyuan County, Jiangxi Province, demonstrate that the HD-TGCN outperforms the state-of-the-art flood prediction models in mean absolute error, Nash–Sutcliffe efficiency, and root-mean-square error, with improvements of 80.32%, 0.15%, and 73.99%, respectively, providing a more accurate flood forecasting method that will play a critical role in future flood disaster prevention and control.

With the increasing interest in the south polar region of the Moon, there is an urgent need for high-resolution mapping products to support future exploration. Bundle block adjustment (BBA) is able to improve the spatial positioning accuracy of images used to generate high-precision mapping products. However, the impacts of weak-convergence geometry and poor illumination conditions in lunar south pole on BBA still need to be solved. This article proposes a large-scale robust BBA method for narrow angle camera (NAC) images in south pole. The NAC images are taken from the Lunar Reconnaissance Orbiter, which are scanner-type images requiring special treatment beyond the classical BBA of framing camera images. To handle the weak-convergence geometry issue of stereo NAC imagery, a topographic constraint using reference digital elevation model is integrated into BBA with an appropriate weighting scheme to estimate local topographic relief. In addition, a two-stage outlier elimination strategy for BBA with absolute-relative thresholding and iteratively reweighting methods is presented to reject outliers caused by the poor illumination conditions in lunar south pole. Thousands of images are used for experimental tests and lunar orbiter laser altimeter digital elevation models (DEMs) with different resolutions are used as reference data. The satisfactory experimental results demonstrate the effectiveness and reliability of our BBA method. The proposed large-scale BBA method reduces the relative positioning errors in block network to within 1 m, and decreases the inconsistency between NAC images caused by the orientation errors to less than 0.5 pixel, which offers a reliable solution for large-scale controlled mapping.

A limited number of electroencephalography (EEG) channels are useful for neonatal sleep classification, particularly in the Internet of Medical Things (IoMT) field, where compact and lightweight devices are essential to monitoring health effectively. A streamlined and cost-effective IoMT solution can be achieved by utilizing fewer EEG channels, thereby reducing data transmission and device processing requirements. Using only two channels of an EEG device, this study presents a binary and multistage classification of neonatal sleep. The binary classification (sleep vs awake) achieved an accuracy of 87.56%, and a Cohen’s kappa of 74.13%. The quiet sleep ( $Q_{S}$ ) detection accuracy was 95.63%, with a Cohen’s kappa of 83.87%. For the three-stage classification, accuracy was 83.72%, and Cohen’s kappa was 69.73%. With only two channels, these are the highest performance parameters. The focus is on the fusion of features extracted through flexible analytical wavelet transform (FAWT) & discrete wavelet transform (DWT), ensemble-based voting models, and fewer channels. To feed crucial features into the ensemble-based voting model, feature importance, feature selection, and validation mechanisms were used. To design the voting classifier, several machine learning models were used, compared, and optimized. With SelectKBest feature selection, the proposed methodology was found to be the most effective. By using only two channels, this study shows the practicality of classifying neonatal sleep stages.

Gaze estimation, as a technique that reflects individual attention, can be used for disability assistance and assisting physicians in diagnosing diseases such as autism spectrum disorder (ASD), Parkinson’s disease, and attention deficit hyperactivity disorder (ADHD). Various techniques have been proposed for gaze estimation and achieved high resolution. Among these approaches, electrooculography (EOG)-based gaze estimation, as an economical and effective method, offers a promising solution for practical applications. Objective: In this paper, we systematically investigated the possible EOG electrode locations which are spatially distributed around the orbital cavity. Afterward, quantities of informative features to characterize physiological information of eye movement from the temporal-spectral domain are extracted from the seven differential channels. Methods and procedures: To select the optimum channels and relevant features, and eliminate irrelevant information, a heuristical search algorithm (i.e., forward stepwise strategy) is applied. Subsequently, a comparative analysis of the impacts of electrode placement and feature contributions on gaze estimation is evaluated via 6 classic models with 18 subjects. Results: Experimental results showed that the promising performance was achieved both in the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) within a wide gaze that ranges from −50° to +50°. The MAE and RMSE can be improved to 2.80° and 3.74° ultimately, while only using 10 features extracted from 2 channels. Compared with the prevailing EOG-based techniques, the performance improvement of MAE and RMSE range from 0.70° to 5.48° and 0.66° to 5.42°, respectively. Conclusion: We proposed a robust EOG-based gaze estimation approach by systematically investigating the optimal channel/feature combination. The experimental results indicated not only the superiority of the proposed approach but also its potential for clinical application. Clinical and translational impact statement: Accurate gaze estimation is a key step for assisting disabilities and accurate diagnosis of various diseases including ASD, Parkinson’s disease, and ADHD. The proposed approach can accurately estimate the points of gaze via EOG signals, and thus has the potential for various related medical applications.

In this article, a position sensorless control strategy for the five-phase interior permanent magnet synchronous motor (IPMSM) by injecting pseudo-random signal into the third harmonic space is proposed. The conventional sensorless control of five-phase IPMSM based on third harmonic space extracts rotor position signal by injecting high-frequency square voltage signals into the third harmonic space. However, high-frequency square-wave voltage injection (HFSVI) produces a loud audible noise. Different from the conventional HFSVI strategy, the proposed strategy injects simplified pseudo-random fixed-frequency phase-switching (SPRFFPS) signal into the third harmonic space of the five-phase IPMSM to reduce high-frequency audible noise. To verify the effectiveness of the proposed strategy, the power spectral density (PSD) of phase current is analyzed. Also, since high-frequency signal is not injected into the fundamental space, the digital filter in the fundamental space current loop will no longer be needed, which avoids time delay and degradation of control performance. Finally, the correctness of the proposed strategy is verified by experiments.

Recent advances in deep learning have led to increased adoption of convolutional neural networks (CNN) for structural magnetic resonance imaging (sMRI)-based Alzheimer’s disease (AD) detection. AD results in widespread damage to neurons in different brain regions and destroys their connections. However, current CNN-based methods struggle to relate spatially distant information effectively. To solve this problem, we propose a graph reasoning module (GRM), which can be directly incorporated into CNN-based AD detection models to simulate the underlying relationship between different brain regions and boost AD diagnosis performance. Specifically, in GRM, an adaptive graph Transformer (AGT) block is designed to adaptively construct a graph representation based on the feature map given by CNN, a graph convolutional network (GCN) block is adopted to update the graph representation, and a feature map reconstruction (FMR) block is built to convert the learned graph representation to a feature map. Experimental results demonstrate that the insertion of the GRM in the existing AD classification model can increase its balanced accuracy by more than 4.3%. The GRM-embedded model achieves state-of-the-art performance compared with current deep learning-based AD diagnosis methods, with a balanced accuracy of 86.2%.

Using graph neural networks to model recommendation scenarios can effectively capture high-order relationship features between objects, thereby helping the model better handle recommendation problems. However, the over-smoothing phenomenon poses a performance constraint for recommendation algorithms based on graph convolutional node aggregation. In realistic recommendation scenarios that involve social relationships, the imbalance of node degrees can deepen the impact of over-smoothing on recommendation accuracy. To address these issues, we propose an adaptive matrix completion algorithm for collaborative filtering recommendation, which is based on the aggregation rules of relational graph convolutional networks, and introduces jumping knowledge connection for adaptive selection of user-item feature aggregation results of deep graph convolutional networks. And in order to overcome the limitations of existing interlayer aggregation mechanisms, we design a self-attention-based aggregation mechanism to integrate the output of each layer and enhance the generalization ability of the model. In addition, we introduce normalization in the process of data transmission between layers to ensure the distinguishability between nodes. Finally, we conduct experiments on three real recommendation datasets to compare the algorithm’s performance and perform ablation analysis. Our model achieves RMSEs of 0.9058, 0.8346 and 0.7176 on the three datasets respectively. The results show that the recommendation performance of our model achieves a leading level when compared with current state-of-the-art algorithms and verifies the influence of node degree distribution on the recommendation process.

Due to severe inter-beam interference and high cost of conventional full space division multiple access (SDMA)-based multi-user underwater visible light communication (MU-UVLC) systems, partial SDMA has been considered to substantially reduce inter-beam interference and costs by enabling closely located users to share the same light beam. Moreover, non-orthogonal multiple access (NOMA) is generally adopted to multiplex users in the same beam. However, for users with similar channel gains, the use of NOMA may suffer from severe error propagation. To this end, we propose and demonstrate a hybrid SDMA and quasi-orthogonal multiple access (QOMA) scheme for performance enhancement of partial SDMA-based MU-UVLC systems. Simulation and experimental results show that, in two-user UVLC system applying hybrid SDMA/QOMA, the bit error rate (BER) performance of the near user can be substantially improved and a much wider range of effective power allocation ratio can be achieved compared with that using conventional hybrid SDMA/NOMA.

KCNQ2 epileptic encephalopathy is relatively common in early-onset neonatal epileptic encephalopathy and seizure severity varied widely, categorized as drug-sensitive epilepsy and drug-resistant epilepsy. However, in clinical practice, anti-seizure medicines need to be gradually adjusted based on seizure control which undoubtedly increases the economic burden of patients, so further positive anti-seizure regimens depend on whether seizure severity can be predicted in advance. In this paper, we proposed a reliable assessment to differentiate between drug-sensitive epilepsy and drug-resistant epilepsy caused by KCNQ2 pathogenic variants. Based on the electroencephalogram (EEG) and electrooculogram (EOG) signals, twenty-four classical temporal and spectral domain features were extracted and Gradient Boosting Decision Tree (GBDT) was employed to distinguish between patients with drug-sensitive epilepsy and drug-resistant epilepsy. In addition, we also systematically investigated the impact of channel combination and feature combination based on the forward stepwise selection strategy. By employing selected channels and features, the classification accuracy can reach 81.25% with a sensitivity of 57.14% and specificity of 100%. Compared with the state-of-the-art techniques, including the functional network, effective network, and common spatial patterns, the improvement of accuracy ranges from 37.5% to 56.25%, indicating the superiority of our proposed method. Overall, the proposed method may provide a promising tool to distinguish different seizure outcomes of KCNQ2 epileptic encephalopathy.

Obstructive sleep apnea (OSA), one of the most common sleep-related breathing disorders, contributes as a potentially life-threatening disease. In this paper, a wearable functional near-infrared spectroscopy (fNIRS) system for OSA monitoring is proposed. As a non-invasive system that can monitor oxygenation and cerebral hemodynamics, the proposed system is dedicated to mapping the pathogenic characteristics of OSA to dynamic changes in blood oxygen concentration and to constructing an automatic approach for assessing OSA. An algorithm including feature extraction, feature selection, and classification is proposed to signals. Permutation entropy(PE), for quantitative measuring the complexity of time series, is firstly involved to characterize the features of the physiological signals. Subsequently, the principal component analysis (PCA) for feature dimensionality reduction and support vector machine (SVM) algorithm for OSA classification are applied. The proposed method has been validated on a dataset that collected by the wearable system. It includes 40 subjects and composes of normal, and various severity cessation of breathing (e.g., mild, moderate, and severe). Experimental results exhibit that the proposed system can effectively distinguish OSA and non-OSA subjects, with an accuracy of 91.89%. The proposed system is expected to pave the novel perspective for OSA assessment in terms of cerebral hemodynamics.

The surface electromyography (EMG) decomposition techniques provide access to motor neuron activities and have been applied to myoelectric control schemes. However, the current decomposition-based myoelectric control mainly focuses on discrete gestures or single-DoF continuous movements. In this study, we aimed to map the motor unit discharges, which were identified from high-density surface EMG, to the three degrees of freedom (DoFs) wrist movements. The 3-DoF wrist torques and high-density surface EMG signals were recorded concurrently from eight non-disabled subjects. The experimental protocol included single-DoF movements and their various combinations. We decoded the motor unit discharges from the EMG signals using a segment-wise decomposition algorithm. Then the neural features were extracted from motor unit discharges and projected to wrist torques with a multiple linear regression model. We compared the performance of two neural features (twitch model and spike counting) and two training schemes (single-DoF and multi-DoF training). On average, 145 ± 33 motor units were identified from each subject, with a pulse-to-noise ratio of 30.8 ± 4.2 dB. Both neural features exhibited high estimation accuracy of 3-DoF wrist torques, with an average $\text{R}^{{2}}$ of 0.76 ± 0.12 and normalized root mean square error of 11.4 ± 3.1%. These results demonstrated the efficiency of the proposed method in continuous estimation of 3-DoF wrist torques, which has the potential to advance dexterous myoelectric control based on neural information.

Most existing neonatal sleep staging appro- aches applied multiple EEG channels to obtain good performance. However, it potentially increased the computational complexity and led to an increased risk of skin disruption to neonates during data acquisition. In this paper, a multi-scale hierarchical neural network (MS-HNN) with a squeeze and excitation (SE) block for neonatal sleep staging is presented in this study on the basis of a single EEG channel. MS-HNN composes of multi-scale convolutional neural network (MSCNN), temporal information learning (TIL) module, and squeeze and excitation (SE) block. MSCNN can extract features from different scales and frequencies, and TIL module can acquire the transition information among adjacent stages. In addition, for these extracted features, SE block can selectively concentrate on informative features and weaken redundant features for achieving better performance. The proposed approach was validated on a clinical dataset involving 64 neonates from the Children’s Hospital of Fudan University (CHFU). The proposed network achieves an accuracy of 75.4% and 76.5% for three-class automatic neonatal sleep staging with the single-EEG channel and the eight-EEG channels, respectively. The experimental results show that the proposed method can maintain good performance by making full use of the information in the single channel while reducing the channels to control the computational overhead.

Sleep staging is the essential step in sleep quality assessment and sleep disorders diagnosis. However, most current automatic sleep staging approaches use recurrent neural networks (RNN), resulting in a relatively large training burden. Moreover, these methods only extract information of the whole epoch or adjacent epochs, ignoring the local signal variations within epoch. To address these issues, a novel deep learning architecture named segmented attention network (SAN) is proposed in this paper. The architecture can be divided into feature extraction (FE) and time sequence encoder (TSE). The FE module consists of multiple multiscale CNN (MMCNN) and residual squeeze and excitation block (SE block). The former extracts features from multiple equal-length EEG segments and the latter reinforced the features. The TSE module based on a multi-head attention mechanism could capture the temporal information in the features extracted by FE module. Noteworthy, in SAN, we replaced the RNN module with a TSE module for temporal learning and made the network faster. The evaluation of the model was performed on two widely used public datasets, Montreal Archive of Sleep Studies (MASS) and Sleep-EDFX, and one clinical dataset from Huashan Hospital of Fudan University, Shanghai, China (HSFU). The proposed model achieved the accuracy of 85.5%, 86.4%, 82.5% on Sleep-EDFX, MASS and HSFU, respectively. The experimental results exhibited favorable performance and consistent improvements of SAN on different datasets in comparison with the state-of-the-art studies. It also proved the necessity of sleep staging by integrating the local characteristics within epochs and adjacent informative features among epochs.

The diagnosis of mild cognitive impairment (MCI), a prodromal stage of Alzheimer’s disease (AD), is essential for initiating timely treatment to delay the onset of AD. Previous studies have shown the potential of functional near-infrared spectroscopy (fNIRS) for diagnosing MCI. However, preprocessing fNIRS measurements requires extensive experience to identify poor-quality segments. Moreover, few studies have explored how proper multi-dimensional fNIRS features influence the classification results of the disease. Thus, this study outlined a streamlined fNIRS preprocessing method to process fNIRS measurements and compared multi-dimensional fNIRS features with neural networks in order to explore how temporal and spatial factors affect the classification of MCI and cognitive normality. More specifically, this study proposed using Bayesian optimization-based auto hyperparameter tuning neural networks to evaluate 1D channel-wise, 2D spatial, and 3D spatiotemporal features of fNIRS measurements for detecting MCI patients. The highest test accuracies of 70.83%, 76.92%, and 80.77% were achieved for 1D, 2D, and 3D features, respectively. Through extensive comparisons, the 3D time-point oxyhemoglobin feature was proven to be a more promising fNIRS feature for detecting MCI by using an fNIRS dataset of 127 participants. Furthermore, this study presented a potential approach for fNIRS data processing, and the designed models required no manual hyperparameter tuning, which promoted the general utilization of fNIRS modality with neural network-based classification to detect MCI.

To protect the information of video stream, many selective video encryption schemes have been proposed based on the H.265/HEVC video. However, most of the existing algorithms are not robust, thus failing to decrypt under packet loss. To further improve the robustness capability of video protection, a robust selective encryption scheme is proposed in this paper. In H.265/HEVC standard, video is encoded into multiple slices, and the slices are decoded independently. Inspired by the feature, each slice is individually encrypted using RC4 stream cipher. The pseudorandom binary sequence (PRBS) for one slice is related to encoding parameters and the SHA-256 hash value of the corresponding slice header, thus ensuring the real-time update of the PRBS and increasing the resistance to chosen-plaintext attack. Two-rounds shifting algorithm is designed to scramble non-zero coefficients of the transform units (TUs) and then motion vector difference (MVD) parameters, quantized transform coefficients (QTCs) and intra prediction modes (IPMs) are selected for encryption. The combination of coefficient scrambling and syntax elements encryption further improves the encryption performance. The proposed scheme is slice-synchronized. Hence, it decrypts normally in the case of packet loss, and supports online real-time interaction. Furthermore, the simulation and analysis results show that the proposed algorithm has format compatibility, high security and low time complexity, which is a promising tool for video cryptographic applications.

Elimination of intra-artifacts in EEG has been overlooked in most of the existing sleep staging systems, especially in deep learning-based approaches. Whether intra-artifacts, originated from the eye movement, chin muscle firing, or heart beating, etc., in EEG signals would lead to a positive or a negative masking effect on deep learning-based sleep staging systems was investigated in this paper. We systematically analyzed several traditional pre-processing methods involving fast Independent Component Analysis (FastICA), Information Maximization (Infomax), and Second-order Blind Source Separation (SOBI). On top of these methods, a SOBI-WT method based on the joint use of the SOBI and Wavelet Transform (WT) is proposed. It offered an effective solution for suppressing artifact components while retaining residual informative data. To provide a comprehensive comparative analysis, these pre-processing methods were applied to eliminate the intra-artifacts and the processed signals were fed to two ready-to-use deep learning models, namely two-step hierarchical neural network (THNN) and SimpleSleepNet for automatic sleep staging. The evaluation was performed on two widely used public datasets, Montreal Archive of Sleep Studies (MASS) and Sleep-EDF Expanded, and a clinical dataset that was collected in Huashan Hospital of Fudan University, Shanghai, China (HSFU). The proposed SOBI-WT method increased the accuracy from 79.0% to 81.3% on MASS, 83.3% to 85.7% on Sleep-EDF Expanded, and 75.5% to 77.1% on HSFU compared with the raw EEG signal, respectively. Experimental results demonstrate that the intra-artifacts bring out a masking negative impact on the deep learning-based sleep staging systems and the proposed SOBI-WT method has the best performance in diminishing this negative impact compared with other artifact elimination methods.

The electroencephalogram (EEG), for measuring the electrophysiological activity of the brain, has been widely applied in automatic detection of epilepsy seizures. Various EEG-based seizure detection algorithms have already yielded high sensitivity, but training those algorithms requires a large amount of labelled data. Data labelling is often done with a lot of human efforts, which is very time-consuming. In this study, we propose a hybrid system integrating an unsupervised learning (UL) module and a supervised learning (SL) module, where the UL module can significantly reduce the workload of data labelling. For preliminary seizure screening, UL synthesizes amplitude-integrated EEG (aEEG) extraction, isolation forest-based anomaly detection, adaptive segmentation, and silhouette coefficient-based anomaly detection evaluation. The UL module serves to quickly locate the determinate subjects (seizure segments and seizure-free segments) and the indeterminate subjects (potential seizure candidates). Afterwards, more robust seizure detection for the indeterminate subjects is performed by the SL using an EasyEnsemble algorithm. EasyEnsemble, as a class-imbalance learning method, can potentially decrease the generalization error of the seizure-free segments. The proposed method can significantly reduce the workload of data labelling while guaranteeing satisfactory performance. The proposed seizure detection system is evaluated using the Children’s Hospital Boston-Massachusetts Institute of Technology (CHB-MIT) scalp EEG dataset, and it achieves a mean accuracy of 92.62%, a mean sensitivity of 95.55%, and a mean specificity of 92.57%. To the best of our knowledge, this is the first epilepsy seizure detection study employing the integration of both the UL and the SL modules, achieving a competitive performance superior or similar to that of the state-of-the-art methods.

Electrical status epilepticus during sleep (ESES) is an epileptic encephalopathy in children with complex clinical manifestations. It is accompanied by specific electroencephalography (EEG) patterns of continuous spike and slow-waves. Quantifying such EEG patterns is critical to the diagnosis of ESES. While most of the existing automatic ESES quantification systems ignore the morphological variations of the signal as well as the individual variability among subjects. To address these issues, this paper presents a hybrid expert system that dedicates to mimicking the decision-making process of clinicians in ESES quantification by taking the morphological variations, individual variability, and medical knowledge into consideration. The proposed hybrid system not only offers a general scheme that could propel a semi-auto morphology analysis-based expert decision model to a fully automated ESES quantification with biogeography-based optimization (BBO), but also proposes a more precise individualized quantification system to involve the personalized characteristics by adopting an individualized parameters-selection framework. The feasibility and reliability of the proposed method are evaluated on a clinical dataset collected from twenty subjects at Children’s Hospital of Fudan University, Shanghai, China. The estimation error for the individualized quantitative descriptor ESES is 0-4.32% and the average estimation error is 0.95% for all subjects. Experimental results show the presented system outperforms existing works and the individualized system significantly improves the performance of ESES quantification. The favorable results indicate that the proposed hybrid expert system for automatic ESES quantification is promising to support the diagnosis of ESES.

Deep sleep staging networks have reached top performance on large-scale datasets. However, these models perform poorer when training and testing on small sleep cohorts due to data inefficiency. Transferring well-trained models from large-scale datasets (source domain) to small sleep cohorts (target domain) is a promising solution but still remains challenging due to the domain-shift issue. In this work, an unsupervised domain adaptation approach, domain statistics alignment (DSA), is developed to bridge the gap between the data distribution of source and target domains. DSA adapts the source models on the target domain by modulating the domain-specific statistics of deep features stored in the Batch Normalization (BN) layers. Furthermore, we have extended DSA by introducing cross-domain statistics in each BN layer to perform DSA adaptively (AdaDSA). The proposed methods merely need the well-trained source model without access to the source data, which may be proprietary and inaccessible. DSA and AdaDSA are universally applicable to various deep sleep staging networks that have BN layers. We have validated the proposed methods by extensive experiments on two state-of-the-art deep sleep staging networks, DeepSleepNet+ and U-time. The performance was evaluated by conducting various transfer tasks on six sleep databases, including two large-scale databases, MASS and SHHS, as the source domain, four small sleep databases as the target domain. Thereinto, clinical sleep records acquired in Huashan Hospital, Shanghai, were used. The results show that both DSA and AdaDSA could significantly improve the performance of source models on target domains, providing novel insights into the domain generalization problem in sleep staging tasks.

Motor function assessment is crucial for post-stroke rehabilitation. Conventional evaluation methods are subjective, heavily depending on the experience of therapists. In light of the strong correlation between the stroke severity level and the performance of activities of daily living (ADLs), we explored the possibility of automatically evaluating the upper-limb Brunnstrom Recovery Stage (BRS) via three typical ADLs (tooth brushing, face washing and drinking). Multimodal data (acceleration, angular velocity, surface electromyography) were synchronously collected from 5 upper-limb-worn sensor modules. The performance of BRS evaluation system is known to be variable with different system parameters (e.g., number of sensor modules, feature types and classifiers). We systematically searched for the optimal parameters from different data segmentation strategies (five window lengths and four overlaps), 42 types of features, 12 feature optimization techniques and 9 classifiers with the leave-one-subject-out cross-validation. To achieve reliable and low-cost monitoring, we further explored whether it was possible to obtain a satisfactory result using a relatively small number of sensor modules. As a result, the proposed approach can correctly recognize the stages of all 27 participants using only three sensor modules with the optimized data segmentation parameters (window length: 7s, overlap: 50%), extracted features (simple square integral, slope sign change, modified mean absolute value 1 and modified mean absolute value 2), the feature optimization method (principal component analysis) and the logistic regression classifier. According to the literature, this is the first study to comprehensively optimize sensor configuration and parameters in each stage of the BRS classification framework. The proposed approach can serve as a factor-screening tool towards the automatic BRS classification and is promising to be further used at home.

Objective. The surface electromyography (EMG) decomposition techniques have shown promising results in neurophysiologic investigations, clinical diagnosis, and human-machine interfacing. However, current decomposition methods could only decode a limited number of motor units (MUs) because of the local convergence. The number of identified MUs remains similar even though more muscles or movements are involved, where multiple motor neuron populations are activated. The objective of this study was to develop a segment-wise decomposition strategy to increase the number of MU decoded from multiple motor neuron populations. Methods. The EMG signals were divided into several segments depending on the number of involved movements. The motor neurons, activated during each movement, were regarded as a population. The convolution kernel compensation (CKC) method was applied individually for each segment to decode the motor unit discharges from each motor neuron population. The MU filters were obtained in each segment and filtrated to estimate the MU spike trains (MUSTs) from the global EMG signals. The decomposition performance was validated on synthetic and experimental EMG signals. Main results. From synthetic EMG signals generated by two motor neuron populations, the proposed segment-wise CKC (swCKC) decoded significantly more MUs during low and medium excitation levels, with an increased rate of 16.3% to 75.4% compared with the conventional CKC. From experimental signals recorded during ten motor tasks, 133±24 MUs with the pulse-to-noise ratio of 36.6±6.5 dB were identified for each subject by swCKC, whereas the conventional CKC identified only 43±12 MUs. Conclusion and Significance. These results indicate the feasibility and superiority of the proposed swCKC to decode MU activities across motor neuron populations, extending the potential applications of EMG decomposition for neural decoding during multiple motor tasks.

Heterogeneous cellular networks (HCNs) play a significant role in 5G networks, but have mainly been modeled and analyzed on a two-dimensional (2D) plane. In this paper, we model a three-dimensional (3D) two-tier HCN consisting of macro-cells and small-cells via a stochastic geometry approach, where the base stations (BSs) in each tier are assigned an appropriate antenna downtilt to enhance the downlink signal and suppress the inter-cell interference. Based on the 3D HCN model, we derive the downlink spatially-averaged coverage probability and area spectral efficiency (ASE) as functions of the BS antenna downtilts, BS heights, BS densities and cell-association bias. To facilitate fast numerical evaluation, we propose accurate approximations for the integral parts in the coverage probability expression. We then obtain the optimal BS antenna downtilts for both tiers that maximize the downlink spatially-averaged coverage probability by using partial derivative and the bisection method. Analytical and simulation results show that for given BS heights, densities of small-cell BSs (SBSs) and cell-association biases of SBSs, our optimized BS antenna downtilts of both tiers can significantly enhance the downlink spatially-averaged coverage probability and ASE in comparison with the fixed BS antenna downtilts network. Useful insights into the 3D deployment of HCNs considering BS antenna downtilts are obtained based on the analytical and simulation results.

In this paper, a novel two-dimensional (2D) generalized optical spatial modulation (GOSM) scheme is proposed for multiple-input multiple-output optical wireless communication (MIMO-OWC) systems. By grouping multiple successive time slots as one time block, 2D GOSM mapping can be performed not only in the space domain but also in the time domain. Specifically, two types of 2D GOSM mapping schemes are designed, including 2D-1 and 2D-2 GOSM mappings. Moreover, to address the high complexity issue of optimal joint maximum-likelihood (ML) detection and the noise amplification and error propagation issues of zero-forcing-based ML (ZF-ML) detection, a deep neural network (DNN)-aided detection scheme is further designed for 2D GOSM systems. Simulation results demonstrate the superiority of the proposed 2D GOSM scheme with deep learning-aided detection for high-speed and low-complexity MIMO-OWC systems. More specifically, a remarkable 3.4-dB signal-to-noise ratio (SNR) gain can be achieved by 2D GOSM in comparison to the conventional one-dimensional (1D) GOSM, when applying the DNN-aided detection.

Due to the mobility of devices, device-to-device (D2D) communication is a promising fifth-generation (5G) technology in dynamic environments for improving message transmission efficiency for group communication. Additionally, all services in an ad hoc network are current Vehicle Ad Hoc Network (VANET) applications. Therefore, D2D communication has been introduced in ad hoc environments to reduce latency during vehicle conversations, such as autonomous vehicle solutions and drone fleet management for cellular vehicle-to-everything (C-V2X) modules and Internet of Drones (IoDs) networks. However, providing secure and effective group communication is an urgent challenge. To solve these problems, we propose a dynamic group management solution based on distributed ledger technology. This study demonstrates that a distributed ledger-based hierarchical architecture for dynamic group management is faster and more adaptable without compromising security and performance. Furthermore, the proposed method can facilitate the transfer of direct communication data without a centralized database, thereby reducing the chance of a single point of failure. In addition, the research was tested by a third party that has established close cooperation with world-leading automotive electronics suppliers in Taiwan.

Sleep staging is the essential step in sleep quality assessment and sleep disorders diagnosis. However, most current automatic sleep staging approaches use recurrent neural networks (RNN), resulting in a relatively large training burden. Moreover, these methods only extract information of the whole epoch or adjacent epochs, ignoring the local signal variations within epoch. To address these issues, a novel deep learning architecture named segmented attention network (SAN) is proposed in this paper. The architecture can be divided into feature extraction (FE) and time sequence encoder (TSE). The FE module consists of multiple multiscale CNN (MMCNN) and residual squeeze and excitation block (SE block). The former extracts features from multiple equal-length EEG segments and the latter reinforced the features. The TSE module based on a multi-head attention mechanism could capture the temporal information in the features extracted by FE module. Noteworthy, in SAN, we replaced the RNN module with a TSE module for temporal learning and made the network faster. The evaluation of the model was performed on two widely used public datasets, Montreal Archive of Sleep Studies (MASS) and Sleep-EDFX, and one clinical dataset from Huashan Hospital of Fudan University, Shanghai, China (HSFU). The proposed model achieved the accuracy of 85.5%, 86.4%, 82.5% on Sleep-EDFX, MASS and HSFU, respectively. The experimental results exhibited favorable performance and consistent improvements of SAN on different datasets in comparison with the state-of-the-art studies. It also proved the necessity of sleep staging by integrating the local characteristics within epochs and adjacent informative features among epochs. [ABSTRACT FROM AUTHOR]

With the emergence of the multicore architecture and the increase of chip operating frequency, power optimization has become a key step of circuit logic synthesis. Aiming at the XNOR/OR circuits, with the goal of minimizing power, construct the optimal polarity fixed-polarity Reed–Muller (FPRM) circuits power optimization scheme. However, the power optimization for FPRM circuits is a multipeak combinatorial optimization problem, we first propose a metaheuristic optimization algorithm (MOA), which includes the global exploration optimizer, local deep exploitation optimizer, and initial population and uses the proposed differential evolution optimization, fierce wolf siege algorithm-based tabu search, and improved skew tent map to make the population evolve. Based on the proposed Huffman tree construction algorithm and MOA, we propose an efficient power optimization approach (EPOA) to find the minimum power FPRM circuit. Experimental results on the benchmark circuits confirm the effectiveness of EPOA. [ABSTRACT FROM AUTHOR]