IEEE Sensors Journal最新文献

{"title":"Depth-Guided Underwater Image Enhancement for Robust Distant-Region Color Correction Under Uneven Illumination","authors":"Fuheng Zhou;Lining Liu;Yulong Huang;Yonggang Zhang","doi":"10.1109/JSEN.2026.3717774","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3717774","url":null,"abstract":"Underwater sensing systems often suffer from uneven illumination and color distortion in distant regions, which reduces the reliability of vision-based perception. Existing underwater image enhancement methods often assume uniform illumination and rely on imperfect reference images, resulting in overenhancement and residual color casts in distant regions that can compromise downstream sensor perception. To address this issue, we propose a depth-guided underwater image correction framework for correcting distant color casts under uneven external illumination. The framework integrates uneven illumination into the underwater image formation model (IFM) and transforms it into three learnable components. These components are combined into a loss function that constrains the framework network during training, enabling it to disentangle image degradation and effectively suppress illumination-induced overenhancement. Furthermore, a depth-guided fusion strategy is introduced to adaptively integrate complementary restoration cues from synthetic-domain and real-world trained models, thereby improving distant-region color correction while preserving near-field color consistency. Experimental results on benchmark underwater datasets demonstrate that the proposed framework achieves superior distant-region color correction, reduces overenhancement artifacts under uneven illumination conditions, and improves the reliability of underwater visual perception.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26575-26590"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ASAMgram: A Novel Approach for the Extraction of Weak Fault Features of Rolling Bearings","authors":"Jiayuan Zhang;Jie Guo;Bingyan Chen;Xiaochi Luan;Tenghui Liu","doi":"10.1109/JSEN.2026.3713486","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3713486","url":null,"abstract":"Early fault features of rolling bearings are typically weak, which makes it challenging for envelope analysis to effectively identify the fault features. To improve the ability to extract early fault features, a novel adaptive spectral amplitude modulation-gram (ASAMgram) method is proposed in this article. The ASAMgram employs a 1/3-binary tree filter bank to decompose the vibration signal into narrowband components. Subsequently, spectral amplitude modulation (SAM) is applied to the narrowband signals, with a magnitude order (MO) range of [0, 1.5]. A discrete normalized squared envelope spectrum peak (DNSESP) index is then defined to adaptively select the optimal MO range for generating the reconstructed maximum squared envelope spectrum (RMSES). Finally, the normalized squared envelope spectrum peak (NSESP) index is established to quantify fault feature prominence within the RMSES, based on which the optimal demodulation frequency band (ODFB) is adaptively selected, thus enhancing the detection of weak fault features. The ASAMgram is verified by the experimental data, and it shows better performance compared with the traditional methods, such as SAM and Fast Kurtogram (FK), in detecting early-stage faults in rolling element bearings.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26257-26266"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust Multimodal Measurement of Chip Thermal Warpage via Fringe Projection Profilometry and Infrared Thermography","authors":"Wanyuan Cai;Haiyang Hou;Yuheng Wang;Wei Tao","doi":"10.1109/JSEN.2026.3711692","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3711692","url":null,"abstract":"Thermal warpage is a critical factor affecting chip reliability evaluation and packaging process optimization. Fringe projection profilometry (FPP) provides dense 3-D point-cloud acquisition and has become an effective approach for warpage characterization. However, chip surface reflectivity variations degrade the reconstruction stability of subtle warpage features. In addition, relying solely on 3-D morphological information makes it difficult to fully characterize the coupling relationship between temperature-field variation and warpage evolution, thereby limiting the multidimensional quantitative analysis of chip thermal warpage. To address these issues, a robust multimodal metrology framework that integrates FPP and infrared thermography is proposed for chip thermal warpage analysis. First, a Gaussian-kernel-optimized robust surface fitting strategy is employed to suppress point-cloud noise, thereby enabling accurate reconstruction of subtle warpage features. Second, the infrared temperature field is mapped onto the reconstructed surface to achieve spatiotemporally consistent characterization of 3-D morphology and temperature-field information. Experimental results demonstrate that the proposed method reduces the plane-fitting deviation from 7.06 to 1.13 <inline-formula> <tex-math>$mu text {m}$ </tex-math></inline-formula> and improves measurement repeatability, with a maximum reduction of 0.146 <inline-formula> <tex-math>$mu text {m}$ </tex-math></inline-formula> in the measurement standard deviation while revealing the coupling relationship between temperature-field variation and warpage evolution. The proposed framework provides an effective solution for robust metrology and multidimensional thermogeometric characterization of chip thermal warpage.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26876-26886"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Method for Generating Small-Sample Data of the Satellite Power Subsystem With InCoTimeGAN Integrating Self-Checking and Information Enhancement","authors":"Jintao Liu;Jun Fan;Guang Jin","doi":"10.1109/JSEN.2026.3707885","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3707885","url":null,"abstract":"Given the strong temporal dependence and high acquisition costs of telemetry data from satellite power subsystems in complex space environments, this study proposes a novel information cooperation time-series generative adversarial network (InCoTimeGAN) model integrated with self-checking and information enhancement capabilities for generating high-fidelity scenario-specific sequential sensor data. Specifically, we optimize the TimeGAN framework by introducing the bidirectional gated recurrent unit (BiGRU), which fully exploits the bidirectional contextual dependencies inherent in time-series sensor data and effectively mitigates the limitation of traditional recurrent networks in capturing long-term dependencies. On this basis, we further integrate channel and spatial attention mechanisms (CSAMs) to adaptively highlight key feature information, thereby significantly improving the fidelity of the generated data. To eliminate abnormal noise and ensure the physical rationality of the generated results, a self-inspection module based on the isolation forest algorithm is proposed. The InCoTimeGAN model is validated on two public datasets (BBVA stock and bearing datasets) and further applied to the small-sample telemetry data generation task of an actual satellite power subsystem—an application scenario closely related to sensor data acquisition and processing. Extensive experimental results demonstrate that it outperforms baseline TimeGAN methods by 71.62% in global maximum mean discrepancy (GMMD), 10.49% in Kolmogorov–Smirnov (KS) divergence, 30.61% in Kullback–Leibler (KL) divergence, and 9.47% in classification accuracy (ACC), respectively. When applied to downstream sensor data-driven tasks, the data generated by InCoTimeGAN reduce the mean squared error (mse) to 0.003, outperforming all baseline models comprehensively. The high-quality synthetic data generated by InCoTimeGAN lay a solid data foundation for reliable data-driven modeling, analysis, and fault diagnosis of satellite power subsystems, which is of great significance for improving the reliability and operational efficiency of spaceborne sensor systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26081-26091"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and Validation of a Wearable Respiratory Monitoring System Based on a 3-D-Printed CB-TPU Strain Sensor","authors":"V. Saroli;A. Addabbo;N. Hussain;S. Maji;E. Schena;C. Massaroni","doi":"10.1109/JSEN.2026.3711835","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3711835","url":null,"abstract":"Respiratory rate (RR) is a key vital sign for assessing physiological status in clinical, occupational, and sports applications. In this context, fused deposition modeling (FDM) offers significant opportunities for the development of wearable respiratory sensors, enabling the fabrication of flexible, low-cost, and body-conformable sensing elements tailored to the human body. This work presents and validates a fully wearable respiratory monitoring system based on a single-layer 3-D-printed carbon-black-filled thermoplastic polyurethane (CB-TPU) strain sensor integrated into a chest strap. The sensing element, previously characterized from a metrological standpoint, was interfaced with dedicated signal-conditioning electronics. Two conditioning strategies, namely a quarter bridge (QB) and a linearizing circuit (LC), were first compared in preliminary tests, leading to the selection of the LC due to its higher sensitivity. The final system was then implemented on a custom printed circuit board (PCB) and integrated into a wearable chest strap. The system was validated on 13 healthy volunteers by comparison with a commercial chest-worn reference device. The protocol included both static conditions (bradypnea, eupnea, and tachypnea) and dynamic activities (walking, jogging, and stair climbing). In the static tests, the mean absolute error (MAE), calculated against the reference device, remained below 2.5 breaths <inline-formula> <tex-math>$cdot $ </tex-math></inline-formula> min⁻¹ across all subjects, with lower errors observed during slower breathing phases. During the dynamic activities, the difference in the mean RR between the proposed system and the reference device remained below 1 breath <inline-formula> <tex-math>$cdot $ </tex-math></inline-formula> min⁻¹. Long-term acquisitions of 42 min further confirmed the robustness during prolonged monitoring and daily life tasks. These results support the proposed system as a promising solution for wearable respiratory monitoring in realistic operating scenarios.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26740-26748"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11611578","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Droplet Viscometry on a Digital Microfluidic Platform via Impedance Sensing","authors":"Yong Le;Yan Huang;Shengyou Zhang;Congwei Liao;Shengdong Zhang","doi":"10.1109/JSEN.2026.3706823","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3706823","url":null,"abstract":"Viscosity is an important transport-related parameter in chemical and bioanalytical systems, yet its measurement in miniaturized platforms often relies on external optical tracking or dedicated sensing structures, which can complicate integration with programmable digital microfluidics (DMFs). Here, we report a calibration-based impedance sensing method for droplet viscosity estimation on an electrowetting-on-dielectric DMF (EWOD-DMF) platform. A small high-frequency excitation is applied to the top plate, while the bottom electrode is used simultaneously for droplet actuation and electrical readout. As the droplet moves across the sensing electrode, the time-varying droplet–electrode overlap produces an impedance response, and a transit-time metric is extracted from the normalized amplitude trajectory. The transit time measured by impedance sensing was validated against imaging-based overlap tracking. Using glycerol–water mixtures with viscosities ranging from 1.01 to 39.23 mPa<inline-formula> <tex-math>$cdot $ </tex-math></inline-formula>s at <inline-formula> <tex-math>$25~^{circ }$ </tex-math></inline-formula>C, we obtained a linear calibration between transit time and viscosity (R<inline-formula> <tex-math>${}^{{2}} =0.99789$ </tex-math></inline-formula>). The method requires neither additional on-chip sensing structures nor camera-based readout during operation, and can be readily integrated with programmable EWOD-DMF workflows. This work demonstrates a calibration-based, low-volume electrical droplet viscosity estimation strategy within the validated range of 1.01–39.23 mPa<inline-formula> <tex-math>$cdot $ </tex-math></inline-formula>s for Newtonian glycerol–water mixtures under the specified EWOD-DMF operating conditions. Extension to complex or non-Newtonian sample systems will require application-specific calibration and validation.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25409-25416"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"YOLO-OODNet: Attention-Guided Oriented Object Detection for Fish-Eye Imaging Sensors","authors":"Kun Zhou;Jiacai Liao;Lin Hu;Jinlai Zhang;Suheng Peng","doi":"10.1109/JSEN.2026.3714195","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3714195","url":null,"abstract":"In object detection tasks, traditional rectangular bounding boxes often include substantial background regions, which can negatively impact detection accuracy. To address this issue, this article adopts oriented object detection, which enables more accurate boundary definition. An improved oriented object detection model, YOLO11-oriented object detection network (YOLO-OODNet), is proposed. The model incorporates a dedicated loss function designed for oriented object detection, which evaluates bounding box regression through a combination of intersection over union (IoU) and confidence information. To further enhance feature representation, we design a new attention module, which leverages a cross-dimensional interaction mechanism to improve the model’s adaptability to spatial structure variations. Furthermore, structural optimizations are applied to the original YOLO11 backbone, enabling improved multiscale object detection while maintaining a lightweight architecture with reduced parameter overhead. In this article, we construct an oriented object detection dataset based on fish-eye image, referred to as the fish-eye Oriented Bounding Box (OBB) dataset. Experimental results demonstrate that YOLO11-OODNet outperforms comparison models across all object categories, which is 3.3% higher than that of the original YOLO11 mAP by 76.5%, while reducing the number of parameters by 258 382, a decrease of 9.7%. Additional evaluations on multiple fish-eye and satellite-based oriented object detection datasets further confirm the proposed model’s superior accuracy and robustness, indicating its strong applicability to scenarios involving complex imaging distortions and pose variations.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"25965-25979"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Practical Acoustic-Based Gesture Recognition Method for Smartphones","authors":"Yipei Fan;Liang Chen;Xiangchen Lu;Shihao Xu;Rising Song;Mao Wang;Guangyi Guo","doi":"10.1109/JSEN.2026.3710238","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3710238","url":null,"abstract":"With the increasing demand for noncontact human–machine interaction, gesture recognition has attracted considerable research interest. Recent works have explored the use of ultrasonic signals for hand gesture tracking and recognition. However, when ultrasonic sensing signals are played concurrently with media audio streams through the same speaker, the superposition can lead to transmitter-side clipping distortion, which degrades the autocorrelation properties of the transmitted signal. In this article, we present an acoustic-based gesture recognition method that addresses such clipping distortion. The proposed approach uses an Android smartphone to transmit modulated ultrasonic signals and employs channel impulse response (CIR) estimates derived from cross correlation to extract gesture-related features. We analyze the effect of transmitter clipping on CIR features and propose a data processing technique to mitigate its impact. Gesture classification is performed using a lightweight convolutional neural network (CNN) model. An Android application was developed, and the method was evaluated under various scenarios. Experimental results demonstrate that, under controlled laboratory conditions with simultaneous media playback, the method achieves recall and precision of at least 95%, and under noisy interference, it maintains recall of at least 85%.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26132-26144"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Attention-Driven Spatiotemporal Deep Learning for Sensor-Based Human Activity Recognition Using Multi-IMU Data","authors":"Mayank Lovanshi;Sanchali Das;Vivek Tiwari","doi":"10.1109/JSEN.2026.3713819","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3713819","url":null,"abstract":"The transition toward the Industry 5.0 paradigm has intensified the need for robust human activity recognition (HAR) systems that enable intelligent human–machine collaboration in privacy-preserving and context-aware environments. Traditional machine learning approaches relying on hand-crafted features often suffer from limited generalizability across users and sensing conditions, while many existing deep learning methods fail to jointly capture intersensor spatial dependencies and long-range temporal dynamics from wearable multi-inertial measurement unit (IMU) data. To address these limitations, this article proposes a unified hybrid deep learning framework for sensor-based HAR that integrates 1-D convolutional neural networks (1D-CNNs), depthwise separable convolutions, multihead self-attention (MHSA) transformers, and temporal convolutional networks (TCNs). The proposed architecture first performs efficient per-IMU local feature extraction using 1D-CNN and depthwise separable convolutions, followed by transformer-based adaptive spatial fusion to model cross-sensor relationships among distributed body-mounted IMUs. Subsequently, TCNs are employed to capture long-range temporal dependencies of human activities, while an attention-based temporal pooling mechanism selectively emphasizes informative temporal representations for robust activity classification. Experimental evaluations conducted on the opportunity and UniMiB-SHAR benchmark datasets demonstrate that the proposed framework consistently outperforms the existing CNN-, RNN-, and transformer-based HAR approaches in terms of accuracy, <inline-formula> <tex-math>$textit {F}1$ </tex-math></inline-formula>-score, and robustness against noisy and missing sensor data. Specifically, the proposed model achieves the improvements of 1.33% in accuracy and 1.24% in <inline-formula> <tex-math>$textit {F}1$ </tex-math></inline-formula>-score on the opportunity dataset and improvements of 1.32% in accuracy and 2.48% in <inline-formula> <tex-math>$textit {F}1$ </tex-math></inline-formula>-score on the UniMiB-SHAR dataset compared with the existing state-of-the-art methods.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26691-26699"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Human Sitting Posture Recognition System Based on Pressure Sensor Layout Optimization and Edge Lightweight AI Inference","authors":"Xiaofei Dong;Huakang Xia;Xinyan Zhou;Lunyao Wang;Yinshui Xia","doi":"10.1109/JSEN.2026.3716712","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3716712","url":null,"abstract":"Human sitting posture recognition is crucial for posture correction, rehabilitation, and health monitoring. However, existing pressure-sensor-based posture recognition systems often face challenges in balancing recognition accuracy and hardware cost. To address these challenges, a human sitting posture recognition system is proposed in this work based on sensor layout optimization and on-device lightweight AI inference. First, a two-stage K-means clustering framework (TKCF) is proposed to identify posture-sensitive regions and optimize the layout and number of pressure sensors. Second, this article designs a lightweight convolutional neural network (LW-CNN) integrated with the squeeze-and-excitation (SE) attention mechanism, which achieves a classification accuracy of 97.96% on the PC platform. After optimizing the layout of sensing points, we train the LW-CNN using the optimized 484-point sensor array and deploy the trained model onto the Microcontroller (MCU). The deployed model occupies just 77.97 kB of flash memory and 18.04 kB of static random access memory (SRAM). In contrast to the original sensor layout, the total quantity of sensing points decreases to 47.27% of the original number, SRAM memory overhead decreases to 47.40% of the baseline value, and the model inference latency decreases to 52.59% of the initial benchmark. These experimental results verify that the proposed method can well balance recognition accuracy, hardware resource cost, and user privacy protection, offering a feasible implementation scheme for lightweight sitting posture recognition and monitoring systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 17","pages":"26935-26947"},"PeriodicalIF":4.5,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148871445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}