Incidence and risk factors of post-operative delirium in glioma patients: A prospective cohort study in general wards.

AIMS: Glioma patients are at high risk for postoperative delirium (POD), yet studies focusing on this population in general neurosurgical ward settings are limited. This paper investigates the incidence of POD and related risk factors in glioma patients hospitalized in general wards. DESIGN: Prospective observational study. METHODS: This prospective study included 133 adult glioma patients hospitalized in the general neurosurgery ward. In addition to collecting routine perioperative general clinical data, patients' psychological status was assessed preoperatively using the Hospital Anxiety and Depression Scale (HADS). POD was assessed within 3 days postoperatively using the Confusion of Consciousness Assessment method, twice daily. The incidence of POD was calculated, and risk factors were identified using logistic regression analysis. RESULTS: The incidence of POD in glioma patients admitted to the general ward was 31.6% (40/133). Multivariate regression revealed advanced age (age > 50 years), frontal lobe tumour, presence of preoperative anxiety or depression, retention of a luminal drain, postoperative pain, indwelling catheter these six factors were independent risk factors for the development of delirium in patients after surgery. CONCLUSION: In general ward settings, supratentorial glioma patients exhibit a high risk of POD. Critical risk factors include preoperative psychological conditions, as well as postoperative pain, drainage and catheterization. Rigorous preoperative evaluations, effective pain management strategies and the integration of humanistic care principles are essential in mitigating the risk of POD for glioma patients. RELEVANCE TO CLINICAL PRACTICE: In general ward settings, this study reveals the high occurrence of POD in glioma patients and identifies preoperative psychological states, age, tumour location and several postoperative factors as significant risk factors for POD, which provides a framework for targeted interventions. By integrating these insights into clinical practice, healthcare teams can better identify glioma patients at risk for POD and implement preventive measures, thereby enhancing recovery and overall care quality for glioma patients in general neurosurgical wards. REPORTING METHOD: This study adheres to the STROBE guidelines, ensuring a transparent and comprehensive reporting of the observational research methodology and results. PATIENT OR PUBLIC CONTRIBUTION: Patients involvement was limited to the provision of data through their participation in the study's assessments and the collection of clinical information. The study did not involve a direct patient or public contribution in the design, conduct, analysis, or interpretation of the data, nor in the preparation of the manuscript.

Rapid and accurate identification of stem cell differentiation stages via SERS and convolutional neural networks.

Monitoring the transition of cell states during induced pluripotent stem cell (iPSC) differentiation is crucial for clinical medicine and basic research. However, both identification category and prediction accuracy need further improvement. Here, we propose a method combining surface-enhanced Raman spectroscopy (SERS) with convolutional neural networks (CNN) to precisely identify and distinguish cell states during stem cell differentiation. First, mitochondria-targeted probes were synthesized by combining AuNRs and mitochondrial localization signal (MLS) peptides to obtain effective and stable SERS spectra signals at various stages of cell differentiation. Then, the SERS spectra served as input datasets, and their distinctive features were learned and distinguished by CNN. As a result, rapid and accurate identification of six different cell states, including the embryoid body (EB) stage, was successfully achieved throughout the stem cell differentiation process with an impressive prediction accuracy of 98.5%. Furthermore, the impact of different spectral feature peaks on the identification results was investigated, which provides a valuable reference for selecting appropriate spectral bands to identify cell states. This is also beneficial for shortening the spectral acquisition region to enhance spectral acquisition speed. These results suggest the potential for SERS-CNN models in quality monitoring of stem cells, advancing the practical applications of stem cells.

SERS-based long-term mitochondrial pH monitoring during differentiation of human induced pluripotent stem cells to neural progenitor cells.

As one of the important organelles in the process of cell differentiation, mitochondria regulate the whole process of differentiation by participating in energy supply and information transmission. Mitochondrial pH value is a key indicator of mitochondrial function. Therefore, real-time monitoring of mitochondrial pH value during cell differentiation is of great significance for understanding cell biochemical processes and exploring differentiation mechanisms. In this study, Surface-enhanced Raman scattering (SERS) technology was used to achieve the real-time monitoring of mitochondrial pH during induced pluripotent stem cells (iPSCs) differentiation into neural progenitor cells (NPCs). The results showed that the variation trend of mitochondrial pH in normal and abnormal differentiated batches was different. The mitochondrial pH value of normal differentiated cells continued to decline from iPSCs to embryoid bodies (EB) day 4, and continued to rise from EB day 4 to the NPCs stage, and the mitochondrial microenvironment of iPSCs to NPCs differentiation became acidic. In contrast, the mitochondrial pH value of abnormally differentiated cells declined continuously during differentiation. This study improves the information on acid-base balance during cell differentiation and may provide a basis for further understanding of the changes and regulatory mechanisms of mitochondrial metabolism during cell differentiation. This also helps to improve more accurate and useful differentiation protocols based on the microenvironment within the mitochondria, improving the efficiency of cell differentiation.

DHM/SERS reveals cellular morphology and molecular changes during iPSCs-derived activation of astrocytes.

The activation of astrocytes derived from induced pluripotent stem cells (iPSCs) is of great significance in neuroscience research, and it is crucial to obtain both cellular morphology and biomolecular information non-destructively in situ, which is still complicated by the traditional optical microscopy and biochemical methods such as immunofluorescence and western blot. In this study, we combined digital holographic microscopy (DHM) and surface-enhanced Raman scattering (SERS) to investigate the activation characteristics of iPSCs-derived astrocytes. It was found that the projected area of activated astrocytes decreased by 67%, while the cell dry mass increased by 23%, and the cells changed from a flat polygonal shape to an elongated star-shaped morphology. SERS analysis further revealed an increase in the intensities of protein spectral peaks (phenylalanine 1001 cm-1, proline 1043 cm-1, etc.) and lipid-related peaks (phosphatidylserine 524 cm-1, triglycerides 1264 cm-1, etc.) decreased in intensity. Principal component analysis-linear discriminant analysis (PCA-LDA) modeling based on spectral data distinguished resting and reactive astrocytes with a high accuracy of 96.5%. The increase in dry mass correlated with the increase in protein content, while the decrease in projected area indicated the adjustment of lipid composition and cell membrane remodeling. Importantly, the results not only reveal the cellular morphology and molecular changes during iPSCs-derived astrocytes activation but also reflect their mapping relationship, thereby providing new insights into diagnosing and treating neurodegenerative diseases.

Fresnel incoherent compressive holography toward 3D videography via dual-channel simultaneous phase-shifting interferometry.

Fresnel incoherent correlation holography (FINCH) enables high-resolution 3D imaging of objects from several 2D holograms under incoherent light and has many attractive applications in motionless 3D fluorescence imaging. However, FINCH has difficulty implementing 3D imaging of dynamic scenes since multiple phase-shifting holograms need to be recorded for removing the bias term and twin image in the reconstructed scene, which requires the object to remain static during this progress. Here, we propose a dual-channel Fresnel noncoherent compressive holography method. First, a pair of holograms with π phase shifts obtained in a single shot are used for removing the bias term noise. Then, a physic-driven compressive sensing (CS) algorithm is used to achieve twin-image-free reconstruction. In addition, we analyze the reconstruction effect and suitability of the CS algorithm and two-step phase-shift filtering algorithm for objects with different complexities. The experimental results show that the proposed method can record hologram videos of 3D dynamic objects and scenes without sacrificing the imaging field of view or resolution. Moreover, the system refocuses images at arbitrary depth positions via computation, hence providing a new method for fast high-throughput incoherent 3D imaging.

Convenient dual-wavelength digital holography based on orthogonal polarization strategy with a Wollaston prism.

Dual-wavelength digital holography effectively expands the measurement range of digital holography, but it increases the complexity of optical system due to non-common-path of two wavelengths. Here, by using orthogonal polarization strategy, we present a dual-wavelength digital holography based on a Wollaston prism (DWDH-WP) to separate the reference beams of two wavelengths and realize the common-path of two wavelengths. A Wollaston prism is inset into the reference beam path of the off-axis digital holography system, so two orthogonal-polarized reference beams of two different wavelengths separated at different directions are generated. Then a dual-wavelength multiplexed interferogram with orthogonal interference fringes is captured by using a monochrome camera, in which both the polarization orientations and the interference fringe orientations of two wavelengths are orthogonal, so the spectral crosstalk of two wavelengths with arbitrary wavelength difference can be avoided. Compared with the existing DWDH method, the proposed DWDH-WP method can conveniently realize the common-path of the reference beams of two wavelengths, so it reveals obvious advantages in spectral separation, spectral crosstalk, system simplification, and adjustment flexibility. Both effectiveness and flexibility of the proposed DWDH-WP method are demonstrated by the phase measurement of the HeLa cell and vortex phase plate.

Adaptive phase-shifting interferometry based on a phase-shifting digital holography algorithm.

Phase-shifting interferometry (PSI) requires accurate phase shifts between interferograms for realizing high-accuracy phase retrieval. This paper presents an adaptive PSI through synchronously capturing phase shifts measurement interferograms and phase measurement interferograms, in which the former is a series of spatial carrier frequency phase-shifting interferograms generated by an additional assembly and the phase shifts are calculated with the single-spectrum phase shifts measurement algorithm (SS-PSMA), the latter is employed for phase retrieval with an adaptive phase-shifting digital holography algorithm (PSDHA) based on complex amplitude recovery. In addition to exhibiting excellent reliability, high-accuracy phase retrieval (0.02 rad), and short calculation time (<25 ms), the proposed adaptive PSDHA is suitable for various interferograms with different fringe shapes and numbers. Importantly, both simulation analysis and experimental result demonstrate that this adaptive PSI based on PSDHA can effectively eliminate phase-shifting errors caused by phase shifter and external disturbance, ensuring high-accuracy phase shifts measurement and phase retrieval, meanwhile significantly reducing phase-shifting interferograms acquisition time and phase retrieval calculation time.

Enhancing scanning electron microscopy imaging quality of weakly conductive samples through unsupervised learning.

Scanning electron microscopy (SEM) is a crucial tool for analyzing submicron-scale structures. However, the attainment of high-quality SEM images is contingent upon the high conductivity of the material due to constraints imposed by its imaging principles. For weakly conductive materials or structures induced by intrinsic properties or organic doping, the SEM imaging quality is significantly compromised, thereby impeding the accuracy of subsequent structure-related analyses. Moreover, the unavailability of paired high-low quality images in this context renders the supervised-based image processing methods ineffective in addressing this challenge. Here, an unsupervised method based on Cycle-consistent Generative Adversarial Network (CycleGAN) was proposed to enhance the quality of SEM images for weakly conductive samples. The unsupervised model can perform end-to-end learning using unpaired blurred and clear SEM images from weakly and well-conductive samples, respectively. To address the requirements of material structure analysis, an edge loss function was further introduced to recover finer details in the network-generated images. Various quantitative evaluations substantiate the efficacy of the proposed method in SEM image quality improvement with better performance than the traditional methods. Our framework broadens the application of artificial intelligence in materials analysis, holding significant implications in fields such as materials science and image restoration.

Raman signal optimization based on residual network adaptive focusing.

Due to its high sensitivity and specificity, Micro-Raman spectroscopy has emerged as a vital technique for molecular recognition and identification. As a weakly scattered signal, ensuring the accurate focus of the sample is essential for acquiring high quality Raman spectral signal and its analysis, especially in some complex microenvironments such as intracellular settings. Traditional autofocus methods are often time consuming or necessitate additional hardware, limiting real-time sample observation and device compatibility. Here, we propose an adaptive focusing method based on residual network to realize rapid and accurate focusing on Micro-Raman measurements. Using only a bright field image of the sample acquired on any image plane, we can predict the defocus distance with a residual network trained by Resnet50, in which the focus position is determined by combining the gradient and discrete cosine transform. Further, detailed regional division of the bright field map used for characterizing the height variation of actual sample surface is performed. As a result, a focus prediction map with 1µm accuracy is obtained from a bright field image in 120 ms. Based on this method, we successfully realize Raman signal optimization and the necessary correction of spectral information. This adaptive focusing method based on residual network is beneficial to further enhance the sensitivity and accuracy of Micro-Raman spectroscopy technology, which is of great significance in promoting the wide application of Raman spectroscopy.

Dual-Domain Reconstruction Network Incorporating Multi-Level Wavelet Transform and Recurrent Convolution for Sparse View Computed Tomography Imaging.

Sparse view computed tomography (SVCT) aims to reduce the number of X-ray projection views required for reconstructing the cross-sectional image of an object. While SVCT significantly reduces X-ray radiation dose and speeds up scanning, insufficient projection data give rise to issues such as severe streak artifacts and blurring in reconstructed images, thereby impacting the diagnostic accuracy of CT detection. To address this challenge, a dual-domain reconstruction network incorporating multi-level wavelet transform and recurrent convolution is proposed in this paper. The dual-domain network is composed of a sinogram domain network (SDN) and an image domain network (IDN). Multi-level wavelet transform is employed in both IDN and SDN to decompose sinograms and CT images into distinct frequency components, which are then processed through separate network branches to recover detailed information within their respective frequency bands. To capture global textures, artifacts, and shallow features in sinograms and CT images, a recurrent convolution unit (RCU) based on convolutional long and short-term memory (Conv-LSTM) is designed, which can model their long-range dependencies through recurrent calculation. Additionally, a self-attention-based multi-level frequency feature normalization fusion (MFNF) block is proposed to assist in recovering high-frequency components by aggregating low-frequency components. Finally, an edge loss function based on the Laplacian of Gaussian (LoG) is designed as the regularization term for enhancing the recovery of high-frequency edge structures. The experimental results demonstrate the effectiveness of our approach in reducing artifacts and enhancing the reconstruction of intricate structural details across various sparse views and noise levels. Our method excels in both performance and robustness, as evidenced by its superior outcomes in numerous qualitative and quantitative assessments, surpassing contemporary state-of-the-art CNNs or Transformer-based reconstruction methods.

Polar optical flow network: extracting overlapped holograms of dynamic particles.

Digital holography with lensless in-line setup has been extensively used in particle field measurements. As particle concentration increases, the holograms of dynamic particles locating at different depths tend to superpose with each other with incoherent overlap, hampering effective measurement of individual particles with incomplete information. Drawing inspiration from suborbicular nature of the in-line holographic fringes, in this study, we propose an optical flow method in polar coordinates to mitigate the overlap issue. The approach employs a radial transformer-enhanced network that leverages both the radial and angular characteristics of the polar hologram. Through ablation tests and experimental results, we have demonstrated the effectiveness and superiority of our proposed method.

Tracking intracellular nuclear targeted-chemotherapy of chidamide-loaded Prussian blue nanocarriers by SERS mapping.

The novel histone deacetylase drug chidamide (CHI) has been proven to regulate gene expression associated with oncogenesis via epigenetic mechanisms. However, huge side effects such as non-targeting, poor intracellular accumulation and low nuclear entry efficiency severely restrict its therapeutic efficacy. Dual-targeted nanodrug delivery systems have been proposed as the solution. Herein, we developed a CHI-loaded drug delivery nanosystem based on Prussian blue (PB) nanocarrier, which combines surface-enhanced Raman scattering (SERS) tracking function with cancer cell/nuclear-targeted chemotherapy capability. With the property of background-free SERS mapping, PB nanocarriers can serve as tracking agents to localize intracellular CHI. The incorporation of targeted molecules specifically enhances the cancer cell/nuclear internalization and chemotherapeutic effects of CHI-loaded PB nanocarriers. In vitro cytotoxicity assay clearly shows that the constructed CHI-loaded PB nanocarriers have significant inhibitory on Jurkat cell proliferation. Furthermore, SERS spectral analysis of Jurkat cells incubated with the CHI-loaded PB nanocarriers reveals obvious features of cellular apoptosis: DNA skeleton fragmentation, chromatin depolymerization, histone acetylation, and nucleosome conformation change. Importantly, this CHI-loaded PB nanocarrier will provide a new insight for lymphoblastic leukemia targeted chemotherapy.

Real-time phase imaging with physics-enhanced network and equivariance.

Learning-based phase imaging balances high fidelity and speed. However, supervised training requires unmistakable and large-scale datasets, which are often hard or impossible to obtain. Here, we propose an architecture for real-time phase imaging based on physics-enhanced network and equivariance (PEPI). The measurement consistency and equivariant consistency of physical diffraction images are used to optimize the network parameters and invert the process from a single diffraction pattern. In addition, we propose a regularization method based total variation kernel (TV-K) function constraint to output more texture details and high-frequency information. The results show that PEPI can produce the object phase quickly and accurately, and the proposed learning strategy performs closely to the fully supervised method in the evaluation function. Moreover, the PEPI solution can handle high-frequency details better than the fully supervised method. The reconstruction results validate the robustness and generalization ability of the proposed method. Specially, our results show that PEPI leads to considerable performance improvement on the imaging inverse problem, thereby paving the way for high-precision unsupervised phase imaging.

Single-shot Fresnel incoherent correlation holography via deep learning based phase-shifting technology.

Fresnel incoherent correlation holography (FINCH) realizes non-scanning three-dimension (3D) images using spatial incoherent illumination, but it requires phase-shifting technology to remove the disturbance of the DC term and twin term that appears in the reconstruction field, thus increasing the complexity of the experiment and limits the real-time performance of FINCH. Here, we propose a single-shot Fresnel incoherent correlation holography via deep learning based phase-shifting (FINCH/DLPS) method to realize rapid and high-precision image reconstruction using only a collected interferogram. A phase-shifting network is designed to implement the phase-shifting operation of FINCH. The trained network can conveniently predict two interferograms with the phase shift of 2/3 π and 4/3 π from one input interferogram. Using the conventional three-step phase-shifting algorithm, we can conveniently remove the DC term and twin term of the FINCH reconstruction and obtain high-precision reconstruction through the back propagation algorithm. The Mixed National Institute of Standards and Technology (MNIST) dataset is used to verify the feasibility of the proposed method through experiments. In the test with the MNIST dataset, the reconstruction results demonstrate that in addition to high-precision reconstruction, the proposed FINCH/DLPS method also can effectively retain the 3D information by calibrating the back propagation distance in the case of reducing the complexity of the experiment, further indicating the feasibility and superiority of the proposed FINCH/DLPS method.

γδ T cells recruitment and local proliferation in brain parenchyma benefit anti-neuroinflammation after cerebral microbleeds.

Background: Cerebral microbleeds (CMBs) are an early sign of many neurological disorders and accompanied by local neuroinflammation and brain damage. As important regulators of immune response and neuroinflammation, the biological behavior and role of γδ T cells after CMBs remain largely unknown. Methods: We made a spot injury of microvessel in the somatosensory cortex to mimic the model of CMBs by two-photon laser and in vivo tracked dynamical behaviors of γδ T cells induced by CMBs using TCR-δGFP transgenic mice. Biological features of γδ T cells in the peri-CMBs parenchyma were decoded by flow cytometry and Raman spectra. In wildtype and γδ T cell-deficient mice, neuroinflammation and neurite degeneration in the peri-CMBs cortex were studied by RNAseq, immunostaining and in vivo imaging respectively. Results: After CMBs, γδ T cells in the dural vessels were tracked to cross the meningeal structure and invade the brain parenchyma in a few days, where the division process of γδ T cells were captured. Parenchymal γδ T cells were highly expressed by CXCR6 and CCR6, similar to meningeal γδ T cells, positive for IL-17A and Ki67 (more than 98%), and they contained abundant substances for energy metabolism and nucleic acid synthesis. In γδ T cell-deficient mice, cortical samples showed the upregulation of neuroinflammatory signaling pathways, enhanced glial response and M1 microglial polarization, and earlier neuronal degeneration in the peri-CMBs brain parenchyma compared with wildtype mice. Conclusion: CMBs induce the accumulation and local proliferation of γδ T cells in the brain parenchyma, and γδ T cells exert anti-neuroinflammatory and neuroprotective effects at the early stage of CMBs.

A SERS-Based Dual-Parameter Monitoring Nanoprobe of ROS and PI3K/Akt during Ginsenoside Rg3-Induced Cell Apoptosis.

Both the reactive oxygen species (ROS) level and Phosphatidylinositol 3 Kinase (PI3K) protein content are two crucial parameters for characterizing states of cell apoptosis. Current methods measure these parameters with two different techniques, respectively, which usually lead to evaluation contingency. Ginsenoside Rg3 exhibits an excellent anticancer effect, which is enacted by the Phosphatidylinositol 3 Kinase/Protein Kinase B (PI3K/Akt) pathway involving ROS; however, the precise mechanism that induces cell apoptosis remains unknown. This is due to the lack of information on quantitative intracellular ROS and PI3K. Here, we used a surface-enhanced Raman scattering (SERS)-based boric acid nanoprobe to monitor the intracellular ROS level and phosphatidylinositol-3,4,5-triphosphate (PI(3,4,5)P3) content, which reflects the regulatory effect of the PI3K/Akt pathway. After treatment with ginsenoside Rg3, the PI3K/Akt content first increased and then decreased as the ROS level increased. Moreover, when the ROS level significantly increased, the mitochondrial membrane potential reduced, thus indicating the dynamic regulation effect of intracellular ROS level on the PI3K/Akt pathway. Importantly, in addition to avoiding evaluation contingency, which is caused by measuring the aforementioned parameters with two different techniques, this SERS-based dual-parameter monitoring nanoprobe provides an effective solution for simultaneous ROS level and PI3K content measurements during cell apoptosis. Furthermore, the intracellular ROS level was also able to have a dynamic regulatory effect on the PI3K/Akt pathway, which is essential for studying ROS/PI3K/Akt-pathway-related cell apoptosis and its activation mechanism.

Semiconductor Plasmon Enhanced Upconversion toward a Flexible Temperature Sensor.

Noninvasive and sensitive thermometry is crucial to human health monitoring and applications in disease diagnosis. Despite recent advances in optical temperature detection, the construction of sensitive wearable temperature sensors remains a considerable challenge. Here, a flexible and biocompatible optical temperature sensor is developed by combining plasmonic semiconductor W18O49 enhanced upconversion emission (UCNPs/WO) with flexible poly(lactic acid) (PLA)-based optical fibers. The UCNPs/WO offers highly thermal-sensitive and obviously enhanced dual-wavelength emissions for ratiometric temperature sensing. The PLA polymer endows the sensor with excellent light-transmitting ability for laser excitation and emission collection and high biocompatibility. The fabricated UCNPs/WO-PLA sensor exhibits stable and rapid temperature response in the range 298-368 K, with a high relative sensitivity of 1.53% K-1 and detection limit as low as ±0.4 K. More importantly, this proposed sensor is demonstrated to possess dual function on real-time detection for physiological thermal changes and heat release, exhibiting great potential in wearable health monitoring and biotherapy applications.

SERS-based error calibration of a TMB-H2O2 colorimetric system.

3,3',5,5'-tetramethylbenzidine (TMB)-H2O2 is widely used as an effective colorimetric system, in which the color reaction is implemented with peroxidase-catalyzed TMB oxidation by H2O2 that usually measured UV-vis absorption spectra or Raman spectra. However, its low accuracy significantly limits its application. Blue charge transfer complex (CTC), which is the product of TMB and H2O2 reaction and is used as the basis for partial colorimetric methods, usually causes colorimetric error owing to changes in the UV-vis absorption and Raman spectra during TMB oxidation under various environmental conditions (catalyst type, temperature, H2O2 concentration). Herein, we propose a surface-enhanced Raman spectrum (SERS)-based error calibration method to improve the accuracy of the TMB-H2O2 colorimetric system. It is found that under 633 nm laser excitation, TMB has three Raman peaks at 1189, 1335 and 1609 cm-1 in the single-electron oxidation phase, and these peaks disappear completely in the two-electron oxidation phase. By comparing these Raman peaks, we can conveniently obtain the actual process information during TMB oxidation. Using the proposed method, the accuracy of the TMB-H2O2 colorimetric system improved by more than 15%. Importantly, this SERS-based TMB-H2O2 error calibration method will open a new horizon for enzyme-linked immunosorbent assay (ELISA) and other biomedical applications.

Diffraction-Net: a robust single-shot holography for multi-distance lensless imaging.

Digital holography based on lensless imaging is a developing method adopted in microscopy and micro-scale measurement. To retrieve complex-amplitude on the sample surface, multiple images are required for common reconstruction methods. A promising single-shot approach points to deep learning, which has been used in lensless imaging but suffering from the unsatisfied generalization ability and stability. Here, we propose and construct a diffraction network (Diff-Net) to connect diffraction images at different distances, which breaks through the limitations of physical devices. The Diff-Net based single-shot holography is robust as there is no practical errors between the multiple images. An iterative complex-amplitude retrieval approach based on light transfer function through the Diff-Net generated multiple images is used for complex-amplitude recovery. This process indicates a hybrid-driven method including both physical model and deep learning, and the experimental results demonstrate that the Diff-Net possesses qualified generalization ability for samples with significantly different morphologies.

Deep learning-based method for non-uniform motion-induced error reduction in dynamic microscopic 3D shape measurement.

The non-uniform motion-induced error reduction in dynamic fringe projection profilometry is complex and challenging. Recently, deep learning (DL) has been successfully applied to many complex optical problems with strong nonlinearity and exhibits excellent performance. Inspired by this, a deep learning-based method is developed for non-uniform motion-induced error reduction by taking advantage of the powerful ability of nonlinear fitting. First, a specially designed dataset of motion-induced error reduction is generated for network training by incorporating complex nonlinearity. Then, the corresponding DL-based architecture is proposed and it contains two parts: in the first part, a fringe compensation module is developed as network pre-processing to reduce the phase error caused by fringe discontinuity; in the second part, a deep neural network is employed to extract the high-level features of error distribution and establish a pixel-wise hidden nonlinear mapping between the phase with motion-induced error and the ideal one. Both simulations and real experiments demonstrate the feasibility of the proposed method in dynamic macroscopic measurement.