A deep reinforcement learning algorithm framework for solving multi-objective traveling salesman problem based on feature transformation.

As a special type of multi-objective combinatorial optimization problems (MOCOPs), the multi-objective traveling salesman problem (MOTSP) plays an important role in practical fields such as transportation and robot control. However, due to the complexity of its solution space and the conflicts between different objectives, it is difficult to obtain satisfactory solutions in a short time. This paper proposes an end-to-end algorithm framework for solving MOTSP based on deep reinforcement learning (DRL). By decomposing strategies, solving MOTSP is transformed into solving multiple single-objective optimization subproblems. Through linear transformation, the features of the MOTSP are combined with the weights of the objective function. Subsequently, a modified graph pointer network (GPN) model is used to solve the decomposed subproblems. Compared with the previous DRL model, the proposed algorithm can solve all the subproblems using only one model without adding weight information as input features. Furthermore, our algorithm can output a corresponding solution for each weight, which increases the diversity of solutions. In order to verify the performance of our proposed algorithm, it is compared with four classical evolutionary algorithms and two DRL algorithms on several MOTSP instances. The comparison shows that our proposed algorithm outperforms the compared algorithms both in terms of training time and the quality of the resulting solutions.

General strategy for developing thick-film micro-thermoelectric coolers from material fabrication to device integration.

Micro-thermoelectric coolers are emerging as a promising solution for high-density cooling applications in confined spaces. Unlike thin-film micro-thermoelectric coolers with high cooling flux at the expense of cooling temperature difference due to very short thermoelectric legs, thick-film micro-thermoelectric coolers can achieve better comprehensive cooling performance. However, they still face significant challenges in both material preparation and device integration. Herein, we propose a design strategy which combines Bi2Te3-based thick film prepared by powder direct molding with micro-thermoelectric cooler integrated via phase-change batch transfer. Accurate thickness control and relatively high thermoelectric performance can be achieved for the thick film, and the high-density-integrated thick-film micro-thermoelectric cooler exhibits excellent performance with maximum cooling temperature difference of 40.6 K and maximum cooling flux of 56.5 W·cm-2 at room temperature. The micro-thermoelectric cooler also shows high temperature control accuracy (0.01 K) and reliability (over 30000 cooling cycles). Moreover, the device demonstrates remarkable capacity in power generation with normalized power density up to 214.0 µW · cm-2 · K-2. This study provides a general and scalable route for developing high-performance thick-film micro-thermoelectric cooler, benefiting widespread applications in thermal management of microsystems.

Exosomal YB-1 facilitates ovarian restoration by MALAT1/miR-211-5p/FOXO3 axis.

Premature ovarian failure (POF) affects many adult women less than 40 years of age and leads to infertility. Mesenchymal stem cells-derived small extracellular vesicles (MSCs-sEVs) are attractive candidates for ovarian function restoration and folliculogenesis for POF due to their safety and efficacy, however, the key mediator in MSCs-sEVs that modulates this response and underlying mechanisms remains elusive. Herein, we reported that YB-1 protein was markedly downregulated in vitro and in vivo models of POF induced with H2O2 and CTX respectively, accompanied by granulosa cells (GCs) senescence phenotype. Notably, BMSCs-sEVs transplantation upregulated YB-1, attenuated oxidative damage-induced cellular senescence in GCs, and significantly improved the ovarian function of POF rats, but that was reversed by YB-1 depletion. Moreover, YB-1 showed an obvious decline in serum and GCs in POF patients. Mechanistically, YB-1 as an RNA-binding protein (RBP) physically interacted with a long non-coding RNA, MALAT1, and increased its stability, further, MALAT1 acted as a competing endogenous RNA (ceRNA) to elevate FOXO3 levels by sequestering miR-211-5p to prevent its degradation, leading to repair of ovarian function. In summary, we demonstrated that BMSCs-sEVs improve ovarian function by releasing YB-1, which mediates MALAT1/miR-211-5p/FOXO3 axis regulation, providing a possible therapeutic target for patients with POF.

Downregulation of lysine-specific histone demethylase 1A (KDM1A/LSD1) in medial prefrontal cortex facilitates chronic stress-induced pain and emotional dysfunction in female mice.

Chronic primary pain, characterized by overlapping symptoms of chronic pain, anxiety, and depression, is strongly associated with stress and is particularly prevalent among females. Recent research has convincingly linked epigenetic modifications in the medial prefrontal cortex (mPFC) to chronic pain and chronic stress. However, our understanding of the role of histone demethylation in the mPFC in chronic stress-induced pain remains limited. In this study, we investigated the function of lysine-specific histone demethylase 1A (KDM1A/LSD1) in the context of chronic overlapping pain comorbid with anxiety and depression in female mice. We employed a chronic variable stress model to induce pain hypersensitivity in the face and hindpaws, as well as anxiety-like and depression-like behaviors, in female mice. Our findings revealed that chronic stress led to a downregulation of KDM1A mRNA and protein expression in the mPFC. Notably, overexpressing KDM1A in the mPFC alleviated the pain hypersensitivity, anxiety-like behaviors, and depression-like behaviors in female mice, without affecting basal pain responses or inducing emotional distress. Conversely, conditional knockout of KDM1A in the mPFC exacerbated pain sensitivity and emotional distress specifically in females. In summary, this study highlights the crucial role of KDM1A in the mPFC in modulating chronic stress-induced overlapping pain, anxiety, and depression in females. Our findings suggest that KDM1A may serve as a potential therapeutic target for treating chronic stress-related overlap pain and associated negative emotional disorders.

Gut microbiome and metabolome profiling in Framingham heart study reveals cholesterol-metabolizing bacteria.

Accumulating evidence suggests that cardiovascular disease (CVD) is associated with an altered gut microbiome. Our understanding of the underlying mechanisms has been hindered by lack of matched multi-omic data with diagnostic biomarkers. To comprehensively profile gut microbiome contributions to CVD, we generated stool metagenomics and metabolomics from 1,429 Framingham Heart Study participants. We identified blood lipids and cardiovascular health measurements associated with microbiome and metabolome composition. Integrated analysis revealed microbial pathways implicated in CVD, including flavonoid, γ-butyrobetaine, and cholesterol metabolism. Species from the Oscillibacter genus were associated with decreased fecal and plasma cholesterol levels. Using functional prediction and in vitro characterization of multiple representative human gut Oscillibacter isolates, we uncovered conserved cholesterol-metabolizing capabilities, including glycosylation and dehydrogenation. These findings suggest that cholesterol metabolism is a broad property of phylogenetically diverse Oscillibacter spp., with potential benefits for lipid homeostasis and cardiovascular health.

Diverse and asymmetric patterns of single-neuron projectome in regulating interhemispheric connectivity.

The corpus callosum, historically considered primarily for homotopic connections, supports many heterotopic connections, indicating complex interhemispheric connectivity. Understanding this complexity is crucial yet challenging due to diverse cell-specific wiring patterns. Here, we utilized public AAV bulk tracing and single-neuron tracing data to delineate the anatomical connection patterns of mouse brains and conducted wide-field calcium imaging to assess functional connectivity across various brain states in male mice. The single-neuron data uncovered complex and dense interconnected patterns, particularly for interhemispheric-heterotopic connections. We proposed a metric "heterogeneity" to quantify the complexity of the connection patterns. Computational modeling of these patterns suggested that the heterogeneity of upstream projections impacted downstream homotopic functional connectivity. Furthermore, higher heterogeneity observed in interhemispheric-heterotopic projections would cause lower strength but higher stability in functional connectivity than their intrahemispheric counterparts. These findings were corroborated by our wide-field functional imaging data, underscoring the important role of heterotopic-projection heterogeneity in interhemispheric communication.

Inhibition of MFN1 restores tamoxifen-induced apoptosis in resistant cells by disrupting aberrant mitochondrial fusion dynamics.

Tamoxifen (TAM) resistance presents a major clinical obstacle in the management of estrogen-sensitive breast cancer, highlighting the need to understand the underlying mechanisms and potential therapeutic approaches. We showed that dysregulated mitochondrial dynamics were involved in TAM resistance by protecting against mitochondrial apoptosis. The dysregulated mitochondrial dynamics were associated with increased mitochondrial fusion and decreased fission, thus preventing the release of mitochondrial cytochrome c to the cytoplasm following TAM treatment. Dynamin-related GTPase protein mitofusin 1 (MFN1), which promotes fusion, was upregulated in TAM-resistant cells, and high MFN1 expression indicated a poor prognosis in TAM-treated patients. Mitochondrial translocation of MFN1 and interaction between MFN1 and mitofusin 2 (MFN2) were enhanced to promote mitochondrial outer membrane fusion. The interaction of MFN1 and cristae-shaping protein optic atrophy 1 (OPA1) and OPA1 oligomerization were reduced due to augmented OPA1 proteolytic cleavage, and their apoptosis-promoting function was reduced due to cristae remodeling. Furthermore, the interaction of MFN1 and BAK were increased, which restrained BAK activation following TAM treatment. Knockdown or pharmacological inhibition of MFN1 blocked mitochondrial fusion, restored BAK oligomerization and cytochrome c release, and amplified activation of caspase-3/9, thus sensitizing resistant cells to apoptosis and facilitating the therapeutic effects of TAM both in vivo and in vitro. Conversely, overexpression of MFN1 alleviated TAM-induced mitochondrial apoptosis and promoted TAM resistance in sensitive cells. These results revealed that dysregulated mitochondrial dynamics contributes to the development of TAM resistance, suggesting that targeting MFN1-mediated mitochondrial fusion is a promising strategy to circumvent TAM resistance.

A Color- and Geometric-Feature-Based Approach for Denoising Three-Dimensional Cultural Relic Point Clouds.

In the acquisition process of 3D cultural relics, it is common to encounter noise. To facilitate the generation of high-quality 3D models, we propose an approach based on graph signal processing that combines color and geometric features to denoise the point cloud. We divide the 3D point cloud into patches based on self-similarity theory and create an appropriate underlying graph with a Markov property. The features of the vertices in the graph are represented using 3D coordinates, normal vectors, and color. We formulate the point cloud denoising problem as a maximum a posteriori (MAP) estimation problem and use a graph Laplacian regularization (GLR) prior to identifying the most probable noise-free point cloud. In the denoising process, we moderately simplify the 3D point to reduce the running time of the denoising algorithm. The experimental results demonstrate that our proposed approach outperforms five competing methods in both subjective and objective assessments. It requires fewer iterations and exhibits strong robustness, effectively removing noise from the surface of cultural relic point clouds while preserving fine-scale 3D features such as texture and ornamentation. This results in more realistic 3D representations of cultural relics.

Mass Spectrometry-Based Proteomic Analysis of Potential Host Proteins Interacting with GP5 in PRRSV-Infected PAMs.

Porcine reproductive and respiratory syndrome virus (PRRSV) is a typical immunosuppressive virus causing a large economic impact on the swine industry. The structural protein GP5 of PRRSV plays a pivotal role in its pathogenicity and immune evasion. Virus-host interactions play a crucial part in viral replication and immune escape. Therefore, understanding the interactions between GP5 and host proteins are significant for porcine reproductive and respiratory syndrome (PRRS) control. However, the interaction network between GP5 and host proteins in primary porcine alveolar macrophages (PAMs) has not been reported. In this study, 709 GP5-interacting host proteins were identified in primary PAMs by immunoprecipitation coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS). Bioinformatics analysis revealed that these proteins were involved in multiple cellular processes, such as translation, protein transport, and protein stabilization. Subsequently, immunoprecipitation and immunofluorescence assay confirmed that GP5 could interact with antigen processing and presentation pathways related proteins. Finally, we found that GP5 may be a key protein that inhibits the antigen processing and presentation pathway during PRRSV infection. The novel host proteins identified in this study will be the candidates for studying the biological functions of GP5, which will provide new insights into PRRS prevention and vaccine development.

Task sub-type states decoding via group deep bidirectional recurrent neural network.

Decoding brain states under different cognitive tasks from functional magnetic resonance imaging (fMRI) data has attracted great attention in the neuroimaging filed. However, the well-known temporal dependency in fMRI sequences has not been fully exploited in existing studies, due to the limited temporal-modeling capacity of the backbone machine learning algorithms and rigid training sample organization strategies upon which the brain decoding methods are built. To address these limitations, we propose a novel method for fine-grain brain state decoding, namely, group deep bidirectional recurrent neural network (Group-DBRNN) model. We first propose a training sample organization strategy that consists of a group-task sample generation module and a multiple-scale random fragment strategy (MRFS) module to collect training samples that contain rich task-relevant brain activity contrast (i.e., the comparison of neural activity patterns between different tasks) and maintain the temporal dependency. We then develop a novel decoding model by replacing the unidirectional RNNs that are widely used in existing brain state decoding studies with bidirectional stacked RNNs to better capture the temporal dependency, and by introducing a multi-task interaction layer (MTIL) module to effectively model the task-relevant brain activity contrast. Our experimental results on the Human Connectome Project task fMRI dataset (7 tasks consisting of 23 task sub-type states) show that the proposed model achieves an average decoding accuracy of 94.7% over the 23 fine-grain sub-type states. Meanwhile, our extensive interpretations of the intermediate features learned in the proposed model via visualizations and quantitative assessments of their discriminability and inter-subject alignment evidence that the proposed model can effectively capture the temporal dependency and task-relevant contrast.

Multifunctional cross-sensitive magnetic alginate-chitosan-polyethylene oxide nanofiber sensor for human-machine interaction.

Flexible nanofiber membranes are compelling materials for the development of functional multi-mode sensors; however, their essential features such as high cross-sensitivity, reliable stability and signal discrimination capability have rarely been realized simultaneously in one sensor. Here, a novel multi-mode sensor with a nanofiber membrane structure based on multiple interpenetrating networks of bidisperse magnetic particles, sodium alginate (SA), chitosan (CHI) in conjunction with polyethylene oxide hydrogels was prepared in a controllable electrospinning technology. Specifically, the morphology distributions of nanofibers could be regulated by the crosslinking degree of the interpenetrating networks and the spinning process parameters. The incorporation of SA and CHI endowed the sensor with desirable flexibility, ideal biocompatibility and skin-friendly property. Besides, the assembled sensors not only displayed preferable magnetic sensitivity of 0.34 T-1 and reliable stability, but also exhibited favorable cross-sensitivity, quick response time, and long-term durability for over 5000 cycles under various mechanical stimuli. Importantly, the multi-mode stimuli could be discriminated via producing opposite electrical signals. Furthermore, based on the signal distinguishability of the sensor, a wearable Morse code translation system assisted by the machine learning algorithm was demonstrated, enabling a high recognizing accuracy (>99.1 %) for input letters and numbers information. Due to the excellent multifunctional sensing characteristics, we believe that the sensor will have a high potential in wearable soft electronics and human-machine interactions.

Mapping dynamic spatial patterns of brain function with spatial-wise attention.

Objective: Using functional magnetic resonance imaging (fMRI) and deep learning to discover the spatial pattern of brain function, or functional brain networks (FBNs) has been attracted many reseachers. Most existing works focus on static FBNs or dynamic functional connectivity among fixed spatial network nodes, but ignore the potential dynamic/time-varying characteristics of the spatial networks themselves. And most of works based on the assumption of linearity and independence, that oversimplify the relationship between blood-oxygen level dependence signal changes and the heterogeneity of neuronal activity within voxels.Approach: To overcome these problems, we proposed a novel spatial-wise attention (SA) based method called Spatial and Channel-wise Attention Autoencoder (SCAAE) to discover the dynamic FBNs without the assumptions of linearity or independence. The core idea of SCAAE is to apply the SA to generate FBNs directly, relying solely on the spatial information present in fMRI volumes. Specifically, we trained the SCAAE in a self-supervised manner, using the autoencoder to guide the SA to focus on the activation regions. Experimental results show that the SA can generate multiple meaningful FBNs at each fMRI time point, which spatial similarity are close to the FBNs derived by known classical methods, such as independent component analysis.Main results: To validate the generalization of the method, we evaluate the approach on HCP-rest, HCP-task and ADHD-200 dataset. The results demonstrate that SA mechanism can be used to discover time-varying FBNs, and the identified dynamic FBNs over time clearly show the process of time-varying spatial patterns fading in and out.Significance: Thus we provide a novel method to understand human brain better. Code is available athttps://github.com/WhatAboutMyStar/SCAAE.

Method for Estimating Road Impulse Ahead of Vehicles in Urban Environment with Microelectromechanical System Three-Dimensional Sensor.

Most automated vehicles (AVs) are equipped with abundant sensors, which enable AVs to improve ride comfort by sensing road elevation, such as speed bumps. This paper proposes a method for estimating the road impulse features ahead of vehicles in urban environments with microelectromechanical system (MEMS) light detection and ranging (LiDAR). The proposed method deploys a real-time estimation of the vehicle pose to solve the problem of sparse sampling of the LiDAR. Considering the LiDAR error model, the proposed method builds the grid height measurement model by maximum likelihood estimation. Moreover, it incorporates height measurements with the LiDAR error model by the Kalman filter and introduces motion uncertainty to form an elevation weight method by confidence eclipse. In addition, a gate strategy based on the Mahalanobis distance is integrated to handle the sharp changes in elevation. The proposed method is tested in the urban environment. The results demonstrate the effectiveness of our method.

Frequency-specific functional difference between gyri and sulci in naturalistic paradigm fMRI.

Disentangling functional difference between cortical folding patterns of gyri and sulci provides novel insights into the relationship between brain structure and function. Previous studies using resting-state functional magnetic resonance imaging (rsfMRI) have revealed that sulcal signals exhibit stronger high-frequency but weaker low-frequency components compared to gyral ones, suggesting that gyri may serve as functional integration centers while sulci are segregated local processing units. In this study, we utilize naturalistic paradigm fMRI (nfMRI) to explore the functional difference between gyri and sulci as it has proven to record stronger functional integrations compared to rsfMRI. We adopt a convolutional neural network (CNN) to classify gyral and sulcal fMRI signals in the whole brain (the global model) and within functional brain networks (the local models). The frequency-specific difference between gyri and sulci is then inferred from the power spectral density (PSD) profiles of the learned filters in the CNN model. Our experimental results show that nfMRI shows higher gyral-sulcal PSD contrast effect sizes in the global model compared to rsfMRI. In the local models, the effect sizes are either increased or decreased depending on frequency bands and functional complexity of the FBNs. This study highlights the advantages of nfMRI in depicting the functional difference between gyri and sulci, and provides novel insights into unraveling the relationship between brain structure and function.

m6A-modified RIPK4 facilitates proliferation and cisplatin resistance in epithelial ovarian cancer.

BACKGROUND: Cisplatin (DDP)-based chemotherapy is a common chemotherapeutic regimen for the treatment of advanced epithelial ovarian cancer (EOC). However, most patients rapidly develop chemoresistance. N6-methyladenosine (m6A) is a pervasive RNA modification, and its specific role and potential mechanism in the regulation of chemosensitivity in EOC remain unclear. METHODS: The expression of RIPK4 and its clinicopathological impact were evaluated in EOC cohorts. The biological effects of RIPK4 were investigated using in vitro and in vivo models. RNA m6A quantification was used to measure total m6A levels in epithelial ovarian cancer cells. Luciferase reporter, MeRIP-qPCR, RIP-qPCR and actinomycin-D assays were used to investigate RNA/RNA interactions and m6A modification of RIPK4 mRNA. RESULTS: We demonstrated that RIPK4, an upregulated mRNA in EOC, acts as an oncogene in EOC cells by promoting tumor cell proliferation and DDP resistance at the clinical, database, cellular, and animal model levels. Mechanistically, METTL3 facilitates m6A modification, and YTHDF1 recognizes the specific m6A-modified site to prevent RIPK4 RNA degradation and upregulate RIPK4 expression. This induces NF-κB activation, resulting in tumor growth and DDP resistance in vitro and in vivo. CONCLUSIONS: Collectively, the present findings reveal a novel mechanism underlying the induction of DDP resistance by m6A-modified RIPK4, that may contribute to overcoming chemoresistance in EOC.

CLEP: Contrastive Learning for Epileptic Seizure Prediction Using a Spatio-Temporal-Spectral Network.

Seizure prediction of epileptic preictal period through electroencephalogram (EEG) signals is important for clinical epilepsy diagnosis. However, recent deep learning-based methods commonly employ intra-subject training strategy and need sufficient data, which are laborious and time-consuming for a practical system and pose a great challenge for seizure predicting. Besides, multi-domain characterizations, including spatio-temporal-spectral dependencies in an epileptic brain are generally neglected or not considered simultaneously in current approaches, and this insufficiency commonly leads to suboptimal seizure prediction performance. To tackle the above issues, in this paper, we propose Contrastive Learning for Epileptic seizure Prediction (CLEP) using a Spatio-Temporal-Spectral Network (STS-Net). Specifically, the CLEP learns intrinsic epileptic EEG patterns across subjects by contrastive learning. The STS-Net extracts multi-scale temporal and spectral representations under different rhythms from raw EEG signals. Then, a novel triple attention layer (TAL) is employed to construct inter-dimensional interaction among multi-domain features. Moreover, a spatio dynamic graph convolution network (sdGCN) is proposed to dynamically model the spatial relationships between electrodes and aggregate spatial information. The proposed CLEP-STS-Net achieves a sensitivity of 96.7% and a false prediction rate of 0.072/h on the CHB-MIT scalp EEG database. We also validate the proposed method on clinical intracranial EEG (iEEG) database from our Xuanwu Hospital of Capital Medical University, and the predicting system yielded a sensitivity of 95%, a false prediction rate of 0.087/h. The experimental results outperform the state-of-the-art studies which validate the efficacy of our method. Our code is available at https://github.com/LianghuiGuo/CLEP-STS-Net.

The Molecular and Function Characterization of Porcine MID2.

Midline2 (MID2/TRIM1) is a member of the tripartite motif-containing (TRIM) family, which is involved in a wide range of cellular processes. However, fundamental studies on porcine MID2 (pMID2) are still lacking. In this study, we identified and characterized the full length MID2 gene of pig (Sus scrofa). The sequence alignment analysis results showed that pMID2 had an N-terminal RING zinc-finger domain, BBC domain, and C-terminal COS box, FN3 motif, and PRY-SPRY domain that were conserved and similar to those of other vertebrates. Furthermore, pMID2 had the highest expression levels in porcine lung and spleen. Serial deletion and site-directed mutagenesis showed that the putative nuclear factor-κB (NF-κB) binding site may be an essential transcription factor for regulating the transcription expression of pMID2. Furthermore, the immunofluorescence assay indicated that pMID2 presented in the cell membrane and cytoplasm. To further study the functions of pMID2, we identified and determined its potential ability to perceive poly (I:C) and IFN-α stimulation. Stimulation experiments showed pMID2 enhanced poly (I:C)-/IFN-α-induced JAK-STAT signaling pathway, indicating that pMID2 might participate in the immune responses. In conclusion, we systematically and comprehensively analyzed the characterizations and functions of pMID2, which provide valuable information to explore the pMID2 functions in innate immunity. Our findings not only enrich the current knowledge of MID2 in IFN signaling regulation but also offer the basis for future research of pig MID2 gene.

Research on Deep Reinforcement Learning Control Algorithm for Active Suspension Considering Uncertain Time Delay.

The uncertain delay characteristic of actuators is a critical factor that affects the control effectiveness of the active suspension system. Therefore, it is crucial to develop a control algorithm that takes into account this uncertain delay in order to ensure stable control performance. This study presents a novel active suspension control algorithm based on deep reinforcement learning (DRL) that specifically addresses the issue of uncertain delay. In this approach, a twin-delayed deep deterministic policy gradient (TD3) algorithm with system delay is employed to obtain the optimal control policy by iteratively solving the dynamic model of the active suspension system, considering the delay. Furthermore, three different operating conditions were designed for simulation to evaluate the control performance: deterministic delay, semi-regular delay, and uncertain delay. The experimental results demonstrate that the proposed algorithm achieves excellent control performance under various operating conditions. Compared to passive suspension, the optimization of body vertical acceleration is improved by more than 30%, and the proposed algorithm effectively mitigates body vibration in the low frequency range. It consistently maintains a more than 30% improvement in ride comfort optimization even under the most severe operating conditions and at different speeds, demonstrating the algorithm's potential for practical application.

Functional brain network identification and fMRI augmentation using a VAE-GAN framework.

Recently, deep learning models have achieved superior performance for mapping functional brain networks from functional magnetic resonance imaging (fMRI) data compared with traditional methods. However, due to the lack of sufficient data and the high dimensionality of brain volume, deep learning models of fMRI tend to suffer from overfitting. In addition, existing methods rarely studied fMRI data augmentation and its application. To address these issues, we developed a VAE-GAN framework that combined a VAE (variational auto-encoder) with a GAN (generative adversarial net) for functional brain network identification and fMRI augmentation. As a generative model, the VAE-GAN models the distribution of fMRI so that it enables the extraction of more generalized features, and thus relieve the overfitting issue. The VAE-GAN is easier to train on fMRI than a standard GAN since it uses latent variables from VAE to generate fake data rather than relying on random noise that is used in a GAN, and it can generate higher quality of fake data than VAE since the discriminator can promote the training of the generator. In other words, the VAE-GAN inherits the advantages of VAE and GAN and avoids their limitations in modeling of fMRI data. Extensive experiments on task fMRI datasets from HCP have proved the effectiveness and superiority of the proposed VAE-GAN framework for identifying both temporal features and functional brain networks compared with existing models, and the quality of fake data is higher than those from VAE and GAN. The results on resting state fMRI of Attention Deficit Hyperactivity Disorder (ADHD)-200 dataset further demonstrated that the fake data generated by the VAE-GAN can help improve the performance of brain network modeling and ADHD classification.

Monitoring of the glazing state of the pitch polishing lap by an image texture analysis method in full-aperture continuous polishing.

Full-aperture continuous polishing using a pitch lap is one of the key processes in finishing large flat optical elements. The pitch lap has relief lines cutting into the surface to increase its surface roughness and to improve the polishing results. During polishing, the pitch lap surface is glazed due to the transverse rheologic flow and creep of the pitch lap by the element and conditioner, which has significant influence on the polishing results. In this study, an image texture analysis method is proposed to monitor the glazing state of the pitch lap. The images of the pitch lap surface are captured online, and the image textures are analyzed with the gray level co-occurrence matrix (GLCM) method. The experiments revealed that the GLCM eigenvalues of the surface image are strongly correlative with the glazing state of the pitch lap, which has a linear impact on the material removal rate and material removal uniformity. To the best of our knowledge, this study provides a novel and useful method to monitor the surface glazing state of the pitch lap.