Xiaoman Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xiaoman Zhang has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Zhifang Li
Key Laboratory of Optoelectronic Science and Technology for Medicine, Ministry of Education, Fujian Provincial Key Laboratory of Photonics Technology, Fujian Provincial Engineering Technology Research Center of Photoelectric Sensing Application, College of Photonic and Electronic Engineering Fujian Normal University Fuzhou Fujian China
4 shared publications - Hui Li
Fujian Normal University
2 shared publications Hui Lin
Key Laboratory of Optoelectronic Science and Technology for Medicine, Ministry of Education, Fujian Provincial Key Laboratory of Photonics Technology, Fujian Provincial Engineering Technology Research Center of Photoelectric Sensing Application, College of Photonic and Electronic Engineering Fujian Normal University Fuzhou Fujian China
2 shared publications- Shulian Wu
Key Laboratory of Optoelectronic Science and Technology for Medicine, Ministry of Education, Fujian Provincial Key Laboratory of Photonics Technology, Fujian Provincial Engineering Technology Research Center of Photoelectric Sensing Application College of Photonic and Electronic Engineering, Fujian Normal University Fuzhou China
2 shared publications Wangbiao Li
Key Laboratory of Optoelectronic Science and Technology for Medicine, Ministry of Education, Fujian Provincial Key Laboratory of Photonics Technology, Fujian Provincial Engineering Technology Research Center of Photoelectric Sensing Application, College of Photonic and Electronic Engineering Fujian Normal University Fuzhou Fujian China
2 shared publications- Xiaochen Zhang
College of Computer and Cyber Security, Fujian Normal University, Fuzhou, China
2 shared publications
Latest Publications
- Open Access
- 24 August 2026
- Published by SAGE Publications in Advances in Mechanical Engineering
- Vol. 18 (8)
A novel multi-task learning framework for bearing remaining useful life prediction under variable operating conditions: Integrating deep hybrid model and invariant consistency strategy
Current existing rolling bearing remaining useful life (RUL) prediction methods often suffer from performance degradation under variable operating conditions due to spurious correlations between source and target domains. To address this, a multi-task learning framework is proposed. Specifically, this paper constructs a deep hybrid model as the shared feature extractor. Based on the idea of bottom-level feature sharing, we integrate the primary regression task and auxiliary tasks for modeling. First, we adopt invariant consistency learning (ICON) to mitigate spurious correlations between the two domains and extract effective domain-invariant features. Second, to further reduce the adverse effects of spurious correlations, we add white noise to generate augmented samples. This approach expands the sample distribution and improves the model’s adaptability to signals under various operating conditions. Finally, we carry out joint optimization and training of the model by using original samples, augmented samples and the two types of tasks. To validate the effectiveness of the proposed method, experiments are conducted on two rolling bearing datasets and compared with other methods. The results demonstrate that the proposed method is capable of effectively predicting the RUL of bearings under varying operating conditions.
- 24 August 2026
- Published by SAGE Publications in Advances in Mechanical Engineering
- Vol. 18 (8)
Open AccessA novel multi-task learning framework for bearing remaining useful life prediction under variable operating conditions: Integrating deep hybrid model and invariant consistency strategy
Current existing rolling bearing remaining useful life (RUL) prediction methods often suffer from performance degradation under variable operating conditions due to spurious correlations between source and target domains. To address this, a multi-task learning framework is proposed. Specifically, this paper constructs a deep hybrid model as the shared feature extractor. Based on the idea of bottom-level feature sharing, we integrate the primary regression task and auxiliary tasks for modeling. First, we adopt invariant consistency learning (ICON) to mitigate spurious correlations between the two domains and extract effective domain-invariant features. Second, to further reduce the adverse effects of spurious correlations, we add white noise to generate augmented samples. This approach expands the sample distribution and improves the model’s adaptability to signals under various operating conditions. Finally, we carry out joint optimization and training of the model by using original samples, augmented samples and the two types of tasks. To validate the effectiveness of the proposed method, experiments are conducted on two rolling bearing datasets and compared with other methods. The results demonstrate that the proposed method is capable of effectively predicting the RUL of bearings under varying operating conditions.
- Open Access
- 17 June 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (12) , 6151
RUL Prediction of Rotating Machinery: A Multi-Channel Information Fusion Forecasting Framework and GMM Evolution-Based Health Indicator Construction
To address the challenges of complex multi-channel signal coupling and insufficient long-term temporal dependency characterization in remaining useful life (RUL) prediction of rotating machinery, this paper proposes a multivariate time series forecasting framework integrating multi-channel information fusion and a self-attention gated augmentation unit (SGAU). First, a multilayer perceptron (MLP) explicitly models nonlinear coupling among channels; SGAU replaces the conventional feed-forward network in the Transformer encoder, using multi-head self-attention outputs as gating signals to adaptively regulate feature transformation. Second, multi-channel signals are predicted via this framework; high-dimensional feature vectors are extracted to construct multi-channel Gaussian mixture models (GMMs). Third, Jensen–Shannon divergence (JSD) quantifies deviations between the target and initial data clusters; centroid distance evolutionary trajectory is fused with JSD to construct the health indicator (HI). Continuous HI predictions yield the RUL prediction curve. Experiments on a self-designed wind turbine gearbox platform and the XJTU-SY bearing dataset demonstrate that the proposed framework outperforms baseline methods on Mean Square (MS), Root Mean Square (RMS), and Energy metrics, with average error reductions of 6.6% and 12.1% in the horizontal and vertical directions on the gearbox dataset and 20.9% and 32.3% on the bearing dataset, confirming its effectiveness and generalization capability.
- 17 June 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (12) , 6151
Open AccessRUL Prediction of Rotating Machinery: A Multi-Channel Information Fusion Forecasting Framework and GMM Evolution-Based Health Indicator Construction
To address the challenges of complex multi-channel signal coupling and insufficient long-term temporal dependency characterization in remaining useful life (RUL) prediction of rotating machinery, this paper proposes a multivariate time series forecasting framework integrating multi-channel information fusion and a self-attention gated augmentation unit (SGAU). First, a multilayer perceptron (MLP) explicitly models nonlinear coupling among channels; SGAU replaces the conventional feed-forward network in the Transformer encoder, using multi-head self-attention outputs as gating signals to adaptively regulate feature transformation. Second, multi-channel signals are predicted via this framework; high-dimensional feature vectors are extracted to construct multi-channel Gaussian mixture models (GMMs). Third, Jensen–Shannon divergence (JSD) quantifies deviations between the target and initial data clusters; centroid distance evolutionary trajectory is fused with JSD to construct the health indicator (HI). Continuous HI predictions yield the RUL prediction curve. Experiments on a self-designed wind turbine gearbox platform and the XJTU-SY bearing dataset demonstrate that the proposed framework outperforms baseline methods on Mean Square (MS), Root Mean Square (RMS), and Energy metrics, with average error reductions of 6.6% and 12.1% in the horizontal and vertical directions on the gearbox dataset and 20.9% and 32.3% on the bearing dataset, confirming its effectiveness and generalization capability.
- Open Access
- 26 March 2025
- Published by Elsevier in NeuroImage
- Vol. 310, 121151
Ensemble network using oblique coronal MRI for Alzheimer’s disease diagnosis
- 26 March 2025
- Published by Elsevier in NeuroImage
- Vol. 310, 121151
Open AccessEnsemble network using oblique coronal MRI for Alzheimer’s disease diagnosis
- 25 September 2024
- Published by Wiley in Journal of Biophotonics
- Vol. 17 (11) , e202400376
Quantitative Characterization of Zebrafish Caudal Fin Regeneration Based on Mueller Matrix OCT In Vivo
Zebrafish serves as a valuable model for studying tissue regeneration due to their comprehensive regenerative abilities, particularly in bone tissue. In this study, a Mueller matrix optical coherence tomography (OCT) system was applied to monitor the regenerative processes of zebrafish caudal fins in vivo. The analysis focused on evaluating the thickness of the caudal fin tip and the distribution of internal bone tissue during the regenerative process. Subsequently, the effect of ectoine solution on the regeneration process was observed and discussed. Our findings revealed that the caudal fin blastema did not exhibit phase‐induced polarization characteristics in the Mueller matrix OCT images. Statistical analyses indicated that the caudal fins did not fully regenerate to their original state within 21 days. Furthermore, the results suggested that ectoine solution could enhance tissue regeneration. This approach provides a method for quantifying zebrafish caudal fin regeneration and advances observation techniques for biomedical and clinical applications.
- 25 September 2024
- Published by Wiley in Journal of Biophotonics
- Vol. 17 (11) , e202400376
Quantitative Characterization of Zebrafish Caudal Fin Regeneration Based on Mueller Matrix OCT In Vivo
Zebrafish serves as a valuable model for studying tissue regeneration due to their comprehensive regenerative abilities, particularly in bone tissue. In this study, a Mueller matrix optical coherence tomography (OCT) system was applied to monitor the regenerative processes of zebrafish caudal fins in vivo. The analysis focused on evaluating the thickness of the caudal fin tip and the distribution of internal bone tissue during the regenerative process. Subsequently, the effect of ectoine solution on the regeneration process was observed and discussed. Our findings revealed that the caudal fin blastema did not exhibit phase‐induced polarization characteristics in the Mueller matrix OCT images. Statistical analyses indicated that the caudal fins did not fully regenerate to their original state within 21 days. Furthermore, the results suggested that ectoine solution could enhance tissue regeneration. This approach provides a method for quantifying zebrafish caudal fin regeneration and advances observation techniques for biomedical and clinical applications.
- 17 June 2024
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry C
- Vol. 12 (28) , 10718-10725
Modulating interface performance between 2D semiconductor MoSi2N4 and its native high-k dielectric Si3N4
Two-dimensional (2D) transition metal silicon nitrides (MSi2N4: M denotes Mo or W) are promising channel materials for nanoelectronics owing to their attractive structural and electronic properties.
- 17 June 2024
- Published by Royal Society of Chemistry (RSC) in Journal of Materials Chemistry C
- Vol. 12 (28) , 10718-10725
Modulating interface performance between 2D semiconductor MoSi2N4 and its native high-k dielectric Si3N4
Two-dimensional (2D) transition metal silicon nitrides (MSi2N4: M denotes Mo or W) are promising channel materials for nanoelectronics owing to their attractive structural and electronic properties.
- 17 June 2024
- Published by Wiley in Journal of Biophotonics
- Vol. 17 (8) , e202400116
Three‐dimensional morphological characterization of blood droplets during the dynamic coagulation process
In this study, we employed a method integrating optical coherence tomography (OCT) with the U‐Net and Visual Geometry Group (VGG)‐Net frameworks within a convolutional neural network for quantitative characterization of the three dimensional whole blood during the dynamic coagulation process. VGG‐Net architecture for the identification of blood droplets across three distinct coagulation stages including drop, gelation, and coagulation achieves an accuracy of up to 99%. In addition, the U‐Net architecture demonstrated proficiency in effectively segmenting uncoagulated and coagulated portions of whole blood, as well as the background. Notably, parameters such as volume of uncoagulated and coagulated segments of the whole blood were successfully employed for the precise quantification of the coagulation process, which indicates well for the potential of future clinical diagnostics and analyses.
- 17 June 2024
- Published by Wiley in Journal of Biophotonics
- Vol. 17 (8) , e202400116
Three‐dimensional morphological characterization of blood droplets during the dynamic coagulation process
In this study, we employed a method integrating optical coherence tomography (OCT) with the U‐Net and Visual Geometry Group (VGG)‐Net frameworks within a convolutional neural network for quantitative characterization of the three dimensional whole blood during the dynamic coagulation process. VGG‐Net architecture for the identification of blood droplets across three distinct coagulation stages including drop, gelation, and coagulation achieves an accuracy of up to 99%. In addition, the U‐Net architecture demonstrated proficiency in effectively segmenting uncoagulated and coagulated portions of whole blood, as well as the background. Notably, parameters such as volume of uncoagulated and coagulated segments of the whole blood were successfully employed for the precise quantification of the coagulation process, which indicates well for the potential of future clinical diagnostics and analyses.
- Open Access
Pressure-Insensitive Epidermal Thickness of Fingertip Skin for Optical Image Encryption
In this study, an internal fingerprint-guided epidermal thickness of fingertip skin is proposed for optical image encryption based on optical coherence tomography (OCT) combined with U-Net architecture of a convolutional neural network (CNN). The epidermal thickness of fingertip skin is calculated by the distance between the upper and lower boundaries of the epidermal layer in cross-sectional optical coherence tomography (OCT) images, which is segmented using CNN, and the internal fingerprint at the epidermis–dermis junction (DEJ) is extracted based on the maximum intensity projection (MIP) algorithm. The experimental results indicate that the internal fingerprint-guided epidermal thickness is insensitive to pressure due to normal correlation coefficients and the encryption process between epidermal thickness maps of fingertip skin under different pressures. In addition, the result of the numerical simulation demonstrates the feasibility and security of the encryption scheme by structural similarity index matrix (SSIM) analysis between the original image and the recovered image with the correct and error keys decryption, respectively. The robustness is analyzed based on the SSIM value in three aspects: different pressures, noise attacks, and data loss. Key randomness is valid by the gray histograms, and the average correlation coefficients of adjacent pixelated values in three directions and the average entropy were calculated. This study suggests that the epidermal thickness of fingertip skin could be seen as important biometric information for information encryption.
Open AccessPressure-Insensitive Epidermal Thickness of Fingertip Skin for Optical Image Encryption
In this study, an internal fingerprint-guided epidermal thickness of fingertip skin is proposed for optical image encryption based on optical coherence tomography (OCT) combined with U-Net architecture of a convolutional neural network (CNN). The epidermal thickness of fingertip skin is calculated by the distance between the upper and lower boundaries of the epidermal layer in cross-sectional optical coherence tomography (OCT) images, which is segmented using CNN, and the internal fingerprint at the epidermis–dermis junction (DEJ) is extracted based on the maximum intensity projection (MIP) algorithm. The experimental results indicate that the internal fingerprint-guided epidermal thickness is insensitive to pressure due to normal correlation coefficients and the encryption process between epidermal thickness maps of fingertip skin under different pressures. In addition, the result of the numerical simulation demonstrates the feasibility and security of the encryption scheme by structural similarity index matrix (SSIM) analysis between the original image and the recovered image with the correct and error keys decryption, respectively. The robustness is analyzed based on the SSIM value in three aspects: different pressures, noise attacks, and data loss. Key randomness is valid by the gray histograms, and the average correlation coefficients of adjacent pixelated values in three directions and the average entropy were calculated. This study suggests that the epidermal thickness of fingertip skin could be seen as important biometric information for information encryption.
- Open Access
- 1 January 2023
- Published by European Alliance for Innovation n.o.
Research on the Evaluation Model of Physical Asset Management Maturity of China's Inter-provincial Grid Enterprises
A comprehensive evaluation of the physical assets of power grid enterprises is conducive to promoting the management of assets of enterprises and avoiding the loss and waste of state-owned assets. This paper collects the objective statistics of the enterprise and the subjective scoring data of the experts in the index system for the past 5 years, and determines the comprehensive weight of each index through the combination of subjective and objective weights. Grey correlation and topsis method is used to establish a dynamic evaluation model for objective data to capture the trend of the comprehensive value of physical assets, and subjective data is used to establish a maturity model to calculate the maturity level of physical assets. In this way, the results of dynamic assessment and maturity assessment are combined to obtain the results of grid enterprise asset evaluation under multi-dimensional dynamic perspective. Finally, an empirical analysis using a grid data shows that the company should learn from the scientific management experience of 2016, maintain a high level of asset structure and utilization efficiency, and focus on improving the optimal management of asset decommissioning and health.
- 1 January 2023
- Published by European Alliance for Innovation n.o.
Open AccessResearch on the Evaluation Model of Physical Asset Management Maturity of China's Inter-provincial Grid Enterprises
A comprehensive evaluation of the physical assets of power grid enterprises is conducive to promoting the management of assets of enterprises and avoiding the loss and waste of state-owned assets. This paper collects the objective statistics of the enterprise and the subjective scoring data of the experts in the index system for the past 5 years, and determines the comprehensive weight of each index through the combination of subjective and objective weights. Grey correlation and topsis method is used to establish a dynamic evaluation model for objective data to capture the trend of the comprehensive value of physical assets, and subjective data is used to establish a maturity model to calculate the maturity level of physical assets. In this way, the results of dynamic assessment and maturity assessment are combined to obtain the results of grid enterprise asset evaluation under multi-dimensional dynamic perspective. Finally, an empirical analysis using a grid data shows that the company should learn from the scientific management experience of 2016, maintain a high level of asset structure and utilization efficiency, and focus on improving the optimal management of asset decommissioning and health.
- Open Access
- 28 April 2021
- Published by Frontiers Media SA in Frontiers in Chemistry
- Vol. 9
A Solid-State Fluorescence Switch Based on Triphenylethene-Functionalized Dithienylethene With Aggregation-Induced Emission
The development of novel dithienylethene-based fluorescence switches in the aggregated state, and the solid state is highly desirable for potential application in the fields of optoelectronics and photopharmacology. In this contribution, three novel triphenylethene-functionalized dithienylethenes (1–3) have been designed and prepared by appending triphenylethene moieties at one end of dithienylethene unit. Their chemical structures are confirmed by 1H NMR, 13C NMR, and HRMS (ESI). They display good photochromic behaviors with excellent fatigue resistance upon irradiation with UV or visible light in Tetrahydrofuran (THF) solution. Before irradiation with UV light, they exhibit Aggregation Induced Emission (AIE) properties and luminescence behaviors in the solid state. Moreover, upon alternating irradiation with UV/visible light, they display effective fluorescent switching behaviors in the aggregated state and the solid state. The experimental results have been validated by the Density Functional Theory (DFT) calculations. Thus, they can be utilized as novel fluorescence switches integrated in smart, solid-state optoelectronic materials and photopharmacology.
- 28 April 2021
- Published by Frontiers Media SA in Frontiers in Chemistry
- Vol. 9
Open AccessA Solid-State Fluorescence Switch Based on Triphenylethene-Functionalized Dithienylethene With Aggregation-Induced Emission
The development of novel dithienylethene-based fluorescence switches in the aggregated state, and the solid state is highly desirable for potential application in the fields of optoelectronics and photopharmacology. In this contribution, three novel triphenylethene-functionalized dithienylethenes (1–3) have been designed and prepared by appending triphenylethene moieties at one end of dithienylethene unit. Their chemical structures are confirmed by 1H NMR, 13C NMR, and HRMS (ESI). They display good photochromic behaviors with excellent fatigue resistance upon irradiation with UV or visible light in Tetrahydrofuran (THF) solution. Before irradiation with UV light, they exhibit Aggregation Induced Emission (AIE) properties and luminescence behaviors in the solid state. Moreover, upon alternating irradiation with UV/visible light, they display effective fluorescent switching behaviors in the aggregated state and the solid state. The experimental results have been validated by the Density Functional Theory (DFT) calculations. Thus, they can be utilized as novel fluorescence switches integrated in smart, solid-state optoelectronic materials and photopharmacology.
- 1 August 2020
- Published by Elsevier in Chemosphere
- Vol. 253, 126777
Formation and inhibition of polycyclic aromatic hydrocarbons from the gasification of cyanobacterial biomass in supercritical water
- 1 August 2020
- Published by Elsevier in Chemosphere
- Vol. 253, 126777
Formation and inhibition of polycyclic aromatic hydrocarbons from the gasification of cyanobacterial biomass in supercritical water