Publications by authors named "Linlin Shen"

56 Publications

Learning a Model-Driven Variational Network for Deformable Image Registration.

IEEE Trans Med Imaging 2021 Aug 30;PP. Epub 2021 Aug 30.

Data-driven deep learning approaches to image registration can be less accurate than conventional iterative approaches, especially when training data is limited. To address this issue and meanwhile retain the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using a variable splitting optimization scheme, we first convert the image registration problem, established in a generic variational framework, into two sub-problems, one with a point-wise, closed-form solution and the other one being a denoising problem. We then propose two neural layers (i.e. warping layer and intensity consistency layer) to model the analytical solution and a residual U-Net (termed generalized denoising layer) to formulate the denoising problem. Finally, we cascade the three neural layers multiple times to form our VR-Net. Extensive experiments on three (two 2D and one 3D) cardiac magnetic resonance imaging datasets show that VR-Net outperforms state-of-the-art deep learning methods on registration accuracy, whilst maintaining the fast inference speed of deep learning and the data-efficiency of variational models.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2021.3108881DOI Listing
August 2021

Automated tumor proportion score analysis for PD-L1 (22C3) expression in lung squamous cell carcinoma.

Sci Rep 2021 Aug 5;11(1):15907. Epub 2021 Aug 5.

Department of Pathology, Fudan University Shanghai Cancer Center, Shanghai, China.

Programmed cell death ligend-1 (PD-L1) expression by immunohistochemistry (IHC) assays is a predictive marker of anti-PD-1/PD-L1 therapy response. With the popularity of anti-PD-1/PD-L1 inhibitor drugs, quantitative assessment of PD-L1 expression becomes a new labor for pathologists. Manually counting the PD-L1 positive stained tumor cells is an obviously subjective and time-consuming process. In this paper, we developed a new computer aided Automated Tumor Proportion Scoring System (ATPSS) to determine the comparability of image analysis with pathologist scores. A three-stage process was performed using both image processing and deep learning techniques to mimic the actual diagnostic flow of the pathologists. We conducted a multi-reader multi-case study to evaluate the agreement between pathologists and ATPSS. Fifty-one surgically resected lung squamous cell carcinoma were prepared and stained using the Dako PD-L1 (22C3) assay, and six pathologists with different experience levels were involved in this study. The TPS predicted by the proposed model had high and statistically significant correlation with sub-specialty pathologists' scores with Mean Absolute Error (MAE) of 8.65 (95% confidence interval (CI): 6.42-10.90) and Pearson Correlation Coefficient (PCC) of 0.9436 ([Formula: see text]), and the performance on PD-L1 positive cases achieved by our method surpassed that of non-subspecialty and trainee pathologists. Those experimental results indicate that the proposed automated system can be a powerful tool to improve the PD-L1 TPS assessment of pathologists.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-021-95372-1DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8342621PMC
August 2021

WaveCNet: Wavelet Integrated CNNs to Suppress Aliasing Effect for Noise-Robust Image Classification.

IEEE Trans Image Process 2021 10;30:7074-7089. Epub 2021 Aug 10.

Though widely used in image classification, convolutional neural networks (CNNs) are prone to noise interruptions, i.e. the CNN output can be drastically changed by small image noise. To improve the noise robustness, we try to integrate CNNs with wavelet by replacing the common down-sampling (max-pooling, strided-convolution, and average pooling) with discrete wavelet transform (DWT). We firstly propose general DWT and inverse DWT (IDWT) layers applicable to various orthogonal and biorthogonal discrete wavelets like Haar, Daubechies, and Cohen, etc., and then design wavelet integrated CNNs (WaveCNets) by integrating DWT into the commonly used CNNs (VGG, ResNets, and DenseNet). During the down-sampling, WaveCNets apply DWT to decompose the feature maps into the low-frequency and high-frequency components. Containing the main information including the basic object structures, the low-frequency component is transmitted into the following layers to generate robust high-level features. The high-frequency components are dropped to remove most of the data noises. The experimental results show that WaveCNets achieve higher accuracy on ImageNet than various vanilla CNNs. We have also tested the performance of WaveCNets on the noisy version of ImageNet, ImageNet-C and six adversarial attacks, the results suggest that the proposed DWT/IDWT layers could provide better noise-robustness and adversarial robustness. When applying WaveCNets as backbones, the performance of object detectors (i.e., faster R-CNN and RetinaNet) on COCO detection dataset are consistently improved. We believe that suppression of aliasing effect, i.e. separation of low frequency and high frequency information, is the main advantages of our approach. The code of our DWT/IDWT layer and different WaveCNets are available at https://github.com/CVI-SZU/WaveCNet.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2021.3101395DOI Listing
August 2021

Fingerprint Presentation Attack Detector Using Global-Local Model.

IEEE Trans Cybern 2021 Jun 16;PP. Epub 2021 Jun 16.

The vulnerability of automated fingerprint recognition systems (AFRSs) to presentation attacks (PAs) promotes the vigorous development of PA detection (PAD) technology. However, PAD methods have been limited by information loss and poor generalization ability, resulting in new PA materials and fingerprint sensors. This article thus proposes a global-local model-based PAD (RTK-PAD) method to overcome those limitations to some extent. The proposed method consists of three modules, called: 1) the global module; 2) the local module; and 3) the rethinking module. By adopting the cut-out-based global module, a global spoofness score predicted from nonlocal features of the entire fingerprint images can be achieved. While by using the texture in-painting-based local module, a local spoofness score predicted from fingerprint patches is obtained. The two modules are not independent but connected through our proposed rethinking module by localizing two discriminative patches for the local module based on the global spoofness score. Finally, the fusion spoofness score by averaging the global and local spoofness scores is used for PAD. Our experimental results evaluated on LivDet 2017 show that the proposed RTK-PAD can achieve an average classification error (ACE) of 2.28% and a true detection rate (TDR) of 91.19% when the false detection rate (FDR) equals 1.0%, which significantly outperformed the state-of-the-art methods by ~10% in terms of TDR (91.19% versus 80.74%).
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2021.3081764DOI Listing
June 2021

A self-supervised feature-standardization-block for cross-domain lung disease classification.

Methods 2021 May 13. Epub 2021 May 13.

Imaging Department, Fifth People's Hospital of Longgang District, Shenzhen, Guangdong, China.

With the advance of deep learning technology, convolutional neural network (CNN) has been wildly used and achieved the state-of-the-art performances in the area of medical image classification. However, most existing medical image classification methods conduct their experiments on only one public dataset. When applying a well-trained model to a different dataset selected from different sources, the model usually shows large performance degradation and needs to be fine-tuned before it can be applied to the new dataset. The goal of this work is trying to solve the cross-domain image classification problem without using data from target domain. In this work, we designed a self-supervised plug-and-play feature-standardization-block (FSB) which consisting of image normalization (INB), contrast enhancement (CEB) and boundary detection blocks (BDB), to extract cross-domain robust feature maps for deep learning framework, and applied the network for chest x-ray-based lung diseases classification. Three classic deep networks, i.e. VGG, Xception and DenseNet and four chest x-ray lung diseases datasets were employed for evaluating the performance. The experimental result showed that when employing feature-standardization-block, all three networks showed better domain adaption performance. The image normalization, contrast enhancement and boundary detection blocks achieved in average 2%, 2% and 5% accuracy improvement, respectively. By combining all three blocks, feature-standardization-block achieved in average 6% accuracy improvement.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.ymeth.2021.05.007DOI Listing
May 2021

TASK1 and TASK3 in orexin neuron of lateral hypothalamus contribute to respiratory chemoreflex by projecting to nucleus tractus solitarius.

FASEB J 2021 05;35(5):e21532

Division of Nephrology, Zhongshan Hospital, Fudan University, Shanghai, China.

TWIK-related acid-sensitive potassium channels (TASKs)-like current was recorded in orexin neurons in the lateral hypothalamus (LH), which are essential in respiratory chemoreflex. However, the specific mechanism responsible for the pH-sensitivity remains elusive. Thus, we hypothesized that TASKs contribute to respiratory chemoreflex. In the present study, we found that TASK1 and TASK3 were expressed in orexin neurons. Blocking TASKs or microinjecting acid artificial cerebrospinal fluid (ACSF) in the LH stimulated breathing. In contrast, alkaline ACSF inhibited breathing, which was attenuated by blocking TASK1. Damage of orexin neurons attenuated the stimulatory effect on respiration caused by microinjection of acid ACSF (at a pH of 6.5) or TASKs antagonists. The orexinA-positive fiber and orexin type 1 receptor (OX1R) neurons were located in the nucleus tractus solitarius (NTS). The exciting effect of acidosis in the LH on respiration was inhibited by blocking OX1R of the NTS. Taken together, we conclude that orexin neurons sense the extracellular pH change through TASKs and regulate respiration by projecting to the NTS.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1096/fj.202002189RDOI Listing
May 2021

Deep learning based neuronal soma detection and counting for Alzheimer's disease analysis.

Comput Methods Programs Biomed 2021 May 10;203:106023. Epub 2021 Mar 10.

Computer Vision Institute, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, Guangdong, 518060, China; AI Research Center for Medical Image Analysis and Diagnosis, Shenzhen University, Shenzhen 518060, China; Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen University, Shenzhen 518060, China. Electronic address:

Background And Objective: Alzheimer's Disease (AD) is associated with neuronal damage and decrease. Micro-Optical Sectioning Tomography (MOST) provides an approach to acquire high-resolution images for neuron analysis in the whole-brain. Application of this technique to AD mouse brain enables us to investigate neuron changes during the progression of AD pathology. However, how to deal with the huge amount of data becomes the bottleneck.

Methods: Using MOST technology, we acquired 3D whole-brain images of six AD mice, and sampled the imaging data of four regions in each mouse brain for AD progression analysis. To count the number of neurons, we proposed a deep learning based method by detecting neuronal soma in the neuronal images. In our method, the neuronal images were first cut into small cubes, then a Convolutional Neural Network (CNN) classifier was designed to detect the neuronal soma by classifying the cubes into three categories, "soma", "fiber", and "background".

Results: Compared with the manual method and currently available NeuroGPS software, our method demonstrates faster speed and higher accuracy in identifying neurons from the MOST images. By applying our method to various brain regions of 6-month-old and 12-month-old AD mice, we found that the amount of neurons in three brain regions (lateral entorhinal cortex, medial entorhinal cortex, and presubiculum) decreased slightly with the increase of age, which is consistent with the experimental results previously reported.

Conclusion: This paper provides a new method to automatically handle the huge amounts of data and accurately identify neuronal soma from the MOST images. It also provides the potential possibility to construct a whole-brain neuron projection to reveal the impact of AD pathology on mouse brain.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.cmpb.2021.106023DOI Listing
May 2021

One-Class Fingerprint Presentation Attack Detection Using Auto-Encoder Network.

IEEE Trans Image Process 2021 28;30:2394-2407. Epub 2021 Jan 28.

Automated Fingerprint Recognition Systems (AFRSs) have been threatened by Presentation Attack (PA) since its existence. It is thus desirable to develop effective presentation attack detection (PAD) methods. However, the unpredictable PAs make PAD be a challenging problem. This paper proposes a novel One-Class PAD (OCPAD) method for Optical Coherence Technology (OCT) images based fingerprint PA detection. The proposed OCPAD model is learned from a training set only consists of Bonafides (i.e. real fingerprints). The reconstruction error and latent code obtained from the trained auto-encoder network in the proposed model is taken as the basis for the following spoofness score calculation. To get more accurate reconstruction error, we propose an activation map based weighting model to further refine the accuracy of reconstruction error. We test different statistics and distance measures and finally use a decision level fusion to make the final prediction. Our experiments are performed using a dataset with 93200 bonafide scans and 48400 PA scans. The results show that the proposed OCPAD can achieve a True Positive Rate (TPR) of 99.43% when the False Positive Rate (FPR) equals to 10% and a TPR of 96.59% when FPR=5%, which significantly outperformed a feature based approach and a supervised learning based model requiring PAs for training.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TIP.2021.3052341DOI Listing
January 2021

Robust facial landmark detection by cross-order cross-semantic deep network.

Neural Netw 2021 Apr 16;136:233-243. Epub 2020 Nov 16.

College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, 518060, China.

Recently, convolutional neural networks (CNNs)-based facial landmark detection methods have achieved great success. However, most of existing CNN-based facial landmark detection methods have not attempted to activate multiple correlated facial parts and learn different semantic features from them that they can not accurately model the relationships among the local details and can not fully explore more discriminative and fine semantic features, thus they suffer from partial occlusions and large pose variations. To address these problems, we propose a cross-order cross-semantic deep network (CCDN) to boost the semantic features learning for robust facial landmark detection. Specifically, a cross-order two-squeeze multi-excitation (CTM) module is proposed to introduce the cross-order channel correlations for more discriminative representations learning and multiple attention-specific part activation. Moreover, a novel cross-order cross-semantic (COCS) regularizer is designed to drive the network to learn cross-order cross-semantic features from different activation for facial landmark detection. It is interesting to show that by integrating the CTM module and COCS regularizer, the proposed CCDN can effectively activate and learn more fine and complementary cross-order cross-semantic features to improve the accuracy of facial landmark detection under extremely challenging scenarios. Experimental results on challenging benchmark datasets demonstrate the superiority of our CCDN over state-of-the-art facial landmark detection methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neunet.2020.11.001DOI Listing
April 2021

Generalized Embedding Regression: A Framework for Supervised Feature Extraction.

IEEE Trans Neural Netw Learn Syst 2020 Nov 4;PP. Epub 2020 Nov 4.

Sparse discriminative projection learning has attracted much attention due to its good performance in recognition tasks. In this article, a framework called generalized embedding regression (GER) is proposed, which can simultaneously perform low-dimensional embedding and sparse projection learning in a joint objective function with a generalized orthogonal constraint. Moreover, the label information is integrated into the model to preserve the global structure of data, and a rank constraint is imposed on the regression matrix to explore the underlying correlation structure of classes. Theoretical analysis shows that GER can obtain the same or approximate solution as some related methods with special settings. By utilizing this framework as a general platform, we design a novel supervised feature extraction approach called jointly sparse embedding regression (JSER). In JSER, we construct an intrinsic graph to characterize the intraclass similarity and a penalty graph to indicate the interclass separability. Then, the penalty graph Laplacian is used as the constraint matrix in the generalized orthogonal constraint to deal with interclass marginal points. Moreover, the L2,1-norm is imposed on the regression terms for robustness to outliers and data's variations and the regularization term for jointly sparse projection learning, leading to interesting semantic interpretability. An effective iterative algorithm is elaborately designed to solve the optimization problem of JSER. Theoretically, we prove that the subproblem of JSER is essentially an unbalanced Procrustes problem and can be solved iteratively. The convergence of the designed algorithm is also proved. Experimental results on six well-known data sets indicate the competitive performance and latent properties of JSER.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TNNLS.2020.3027602DOI Listing
November 2020

TLR4/MyD88/NF-κB Signaling in the Rostral Ventrolateral Medulla Is Involved in the Depressor Effect of Candesartan in Stress-Induced Hypertensive Rats.

ACS Chem Neurosci 2020 10 21;11(19):2978-2988. Epub 2020 Sep 21.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai 200032, China.

This study aimed to investigate whether the proinflammatory and pressor effects of endogenous angiotensin II (AngII) are mediated by binding to the AngII type 1 receptor (ATR) and subsequently activating central Toll-like receptor 4 (TLR4) in the rostral ventrolateral medulla (RVLM) of stress-induced hypertensive rats (SIHR). The stress-induced hypertension (SIH) model was established by random electric foot shocks combined with noise stimulation. Mean arterial pressure, heart rate, plasma norepinephrine, and RVLM AngII and TLR4 increased in a time-dependent manner in SIHR. Pro-inflammatory cytokines (tumor necrosis factor α (TNF-α), interleukin 1β (IL-1β)), myeloid differentiation factor 88 (MyD88), and nuclear factor (NF)-κB also increased, while anti-inflammatory cytokine IL-10 decreased in the RVLM of SIHR. These changes were attenuated by 14-day intracerebroventricular (ICV) infusion of VIPER (a TLR4 inhibitor) or candesartan (an ATR antagonist). Both TLR4 and ATR were expressed in the neurons and microglia in the RVLM of SIHR. Candesartan attenuated the expression of TLR4 in the RVLM of SIHR. This study demonstrated that endogenous AngII may activate ATR to upregulate TLR4/MyD88/NF-κB signaling and subsequently trigger an inflammatory response in the RVLM of SIHR, which in turn enhanced sympathetic activity and increased blood pressure.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acschemneuro.0c00029DOI Listing
October 2020

Asthmatic Airway Vagal Hypertonia Involves Chloride Dyshomeostasis of Preganglionic Neurons in Rats.

Front Neurosci 2020 31;14:31. Epub 2020 Jan 31.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai, China.

Airway vagal hypertonia is closely related to the severity of asthma; however, the mechanisms of its genesis are unclear. This study aims to prove that asthmatic airway vagal hypertonia involves neuronal Cl dyshomeostasis. The experimental airway allergy model was prepared with ovalbumin in male adult Sprague-Dawley rats. Plethysmography was used to evaluate airway vagal response to intracisternally injected γ-aminobutyric acid (GABA). Immunofluorescent staining and Western-blot assay were used to examine the expression of microglia-specific proteins, Na-K-2Cl co-transporter 1 (NKCC1), K-Cl co-transporter 2 (KCC2) and brain-derived nerve growth factor (BDNF) in airway vagal centers. Pulmonary inflammatory changes were examined with hematoxylin and eosin staining of lung sections and ELISA assay of ovalbumin-specific IgE in bronchoalveolar lavage fluid (BALF). The results showed that histochemically, experimental airway allergy activated microglia, upregulated NKCC1, downregulated KCC2, and increased the content of BDNF in airway vagal centers. Functionally, experimental airway allergy augmented the excitatory airway vagal response to intracisternally injected GABA, which was attenuated by intracisternally pre-injected NKCC1 inhibitor bumetanide. All of the changes induced by experimental airway allergy were prevented or mitigated by chronic intracerebroventricular or intraperitoneal injection of minocycline, an inhibitor of microglia activation. These results demonstrate that experimental airway allergy augments the excitatory response of airway vagal centers to GABA, which might be the result of neuronal Cl dyshomeostasis subsequent to microglia activation, increased BDNF release and altered expression of Cl transporters. Cl dyshomeostasis in airway vagal centers might contribute to the genesis of airway vagal hypertonia in asthma.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fnins.2020.00031DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7005078PMC
January 2020

Attention by Selection: A Deep Selective Attention Approach to Breast Cancer Classification.

IEEE Trans Med Imaging 2020 06 24;39(6):1930-1941. Epub 2019 Dec 24.

Deep learning approaches are widely applied to histopathological image analysis due to the impressive levels of performance achieved. However, when dealing with high-resolution histopathological images, utilizing the original image as input to the deep learning model is computationally expensive, while resizing the original image to achieve low resolution incurs information loss. Some hard-attention based approaches have emerged to select possible lesion regions from images to avoid processing the original image. However, these hard-attention based approaches usually take a long time to converge with weak guidance, and valueless patches may be trained by the classifier. To overcome this problem, we propose a deep selective attention approach that aims to select valuable regions in the original images for classification. In our approach, a decision network is developed to decide where to crop and whether the cropped patch is necessary for classification. These selected patches are then trained by the classification network, which then provides feedback to the decision network to update its selection policy. With such a co-evolution training strategy, we show that our approach can achieve a fast convergence rate and high classification accuracy. Our approach is evaluated on a public breast cancer histopathological image database, where it demonstrates superior performance compared to state-of-the-art deep learning approaches, achieving approximately 98% classification accuracy while only taking 50% of the training time of the previous hard-attention approach.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2962013DOI Listing
June 2020

Multi-resolution convolutional networks for chest X-ray radiograph based lung nodule detection.

Artif Intell Med 2020 03 28;103:101744. Epub 2019 Oct 28.

Imaging Department of Shenzhen University Health Science Center, Shenzhen University School of Medicine, Shenzhen Second People's Hospital, First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, PR China. Electronic address:

Lung cancer is the leading cause of cancer death worldwide. Early detection of lung cancer is helpful to provide the best possible clinical treatment for patients. Due to the limited number of radiologist and the huge number of chest x-ray radiographs (CXR) available for observation, a computer-aided detection scheme should be developed to assist radiologists in decision-making. While deep learning showed state-of-the-art performance in several computer vision applications, it has not been used for lung nodule detection on CXR. In this paper, a deep learning-based lung nodule detection method was proposed. We employed patch-based multi-resolution convolutional networks to extract the features and employed four different fusion methods for classification. The proposed method shows much better performance and is much more robust than those previously reported researches. For publicly available Japanese Society of Radiological Technology (JSRT) database, more than 99% of lung nodules can be detected when the false positives per image (FPs/image) was 0.2. The FAUC and R-CPM of the proposed method were 0.982 and 0.987, respectively. The proposed approach has the potential of applications in clinical practice.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.artmed.2019.101744DOI Listing
March 2020

A Multi-Organ Nucleus Segmentation Challenge.

IEEE Trans Med Imaging 2020 05 23;39(5):1380-1391. Epub 2019 Oct 23.

Generalized nucleus segmentation techniques can contribute greatly to reducing the time to develop and validate visual biomarkers for new digital pathology datasets. We summarize the results of MoNuSeg 2018 Challenge whose objective was to develop generalizable nuclei segmentation techniques in digital pathology. The challenge was an official satellite event of the MICCAI 2018 conference in which 32 teams with more than 80 participants from geographically diverse institutes participated. Contestants were given a training set with 30 images from seven organs with annotations of 21,623 individual nuclei. A test dataset with 14 images taken from seven organs, including two organs that did not appear in the training set was released without annotations. Entries were evaluated based on average aggregated Jaccard index (AJI) on the test set to prioritize accurate instance segmentation as opposed to mere semantic segmentation. More than half the teams that completed the challenge outperformed a previous baseline. Among the trends observed that contributed to increased accuracy were the use of color normalization as well as heavy data augmentation. Additionally, fully convolutional networks inspired by variants of U-Net, FCN, and Mask-RCNN were popularly used, typically based on ResNet or VGG base architectures. Watershed segmentation on predicted semantic segmentation maps was a popular post-processing strategy. Several of the top techniques compared favorably to an individual human annotator and can be used with confidence for nuclear morphometrics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2947628DOI Listing
May 2020

Three dimensional convolutional neural network-based classification of conduct disorder with structural MRI.

Brain Imaging Behav 2020 Dec;14(6):2333-2340

Computer Vision Institute, School of Computer Science and Software Engineering, Shenzhen University, Shenzhen, People's Republic of China.

Conduct disorder (CD) is a common child and adolescent psychiatric disorder with various representative symptoms, and may cause long-term burden to patients and society. Recently, an increasing number of studies have used deep learning-based approaches, such as convolutional neural network (CNN), to analyze neuroimaging data and to identify biomarkers. In this study, we applied an optimized 3D AlexNet CNN model to automatically extract multi-layer high dimensional features of structural magnetic resonance imaging (sMRI), and to classify CD from healthy controls (HCs). We acquired high-resolution sMRI from 60 CD and 60 age- and gender-matched HCs. All subjects were male, and the age (mean ± std. dev) of participants in the CD and HC groups was 15.3 ± 1.0 and 15.5 ± 0.7, respectively. Five-fold cross validation (CV) was used to train and test this model. The receiver operating characteristic (ROC) curve for this model and that for support vector machine (SVM) model were compared. Feature visualization was performed to obtain intuition about the sMRI features learned by our AlexNet model. Our proposed AlexNet model achieved high classification performance with accuracy of 0.85, specificity of 0.82 and sensitivity of 0.87. The area under the ROC curve (AUC) of AlexNet was 0.86, significantly higher than that of SVM (AUC = 0.78; p = 0.046). The saliency maps for each convolutional layer highlighted the different brain regions in sMRI of CD, mainly including the frontal lobe, superior temporal gyrus, parietal lobe and occipital lobe. The classification results indicated that deep learning-based method is able to explore the hidden features from the sMRI of CD and might assist clinicians in the diagnosis of CD.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11682-019-00186-5DOI Listing
December 2020

Reverse active learning based atrous DenseNet for pathological image classification.

BMC Bioinformatics 2019 Aug 28;20(1):445. Epub 2019 Aug 28.

The Sixth People's Hospital of Shenzhen, Shenzhen, China.

Background: Due to the recent advances in deep learning, this model attracted researchers who have applied it to medical image analysis. However, pathological image analysis based on deep learning networks faces a number of challenges, such as the high resolution (gigapixel) of pathological images and the lack of annotation capabilities. To address these challenges, we propose a training strategy called deep-reverse active learning (DRAL) and atrous DenseNet (ADN) for pathological image classification. The proposed DRAL can improve the classification accuracy of widely used deep learning networks such as VGG-16 and ResNet by removing mislabeled patches in the training set. As the size of a cancer area varies widely in pathological images, the proposed ADN integrates the atrous convolutions with the dense block for multiscale feature extraction.

Results: The proposed DRAL and ADN are evaluated using the following three pathological datasets: BACH, CCG, and UCSB. The experiment results demonstrate the excellent performance of the proposed DRAL + ADN framework, achieving patch-level average classification accuracies (ACA) of 94.10%, 92.05% and 97.63% on the BACH, CCG, and UCSB validation sets, respectively.

Conclusions: The DRAL + ADN framework is a potential candidate for boosting the performance of deep learning models for partially mislabeled training datasets.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12859-019-2979-yDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6712615PMC
August 2019

Adaptive Weighting of Handcrafted Feature Losses for Facial Expression Recognition.

IEEE Trans Cybern 2021 May 15;51(5):2787-2800. Epub 2021 Apr 15.

Due to the importance of facial expressions in human-machine interaction, a number of handcrafted features and deep neural networks have been developed for facial expression recognition. While a few studies have shown the similarity between the handcrafted features and the features learned by deep network, a new feature loss is proposed to use feature bias constraint of handcrafted and deep features to guide the deep feature learning during the early training of network. The feature maps learned with and without the proposed feature loss for a toy network suggest that our approach can fully explore the complementarity between handcrafted features and deep features. Based on the feature loss, a general framework for embedding the traditional feature information into deep network training was developed and tested using the FER2013, CK+, Oulu-CASIA, and MMI datasets. Moreover, adaptive loss weighting strategies are proposed to balance the influence of different losses for different expression databases. The experimental results show that the proposed feature loss with adaptive weighting achieves much better accuracy than the original handcrafted feature and the network trained without using our feature loss. Meanwhile, the feature loss with adaptive weighting can provide complementary information to compensate for the deficiency of a single feature.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2019.2925095DOI Listing
May 2021

3D Neuron Reconstruction in Tangled Neuronal Image With Deep Networks.

IEEE Trans Med Imaging 2020 02 9;39(2):425-435. Epub 2019 Jul 9.

Digital reconstruction or tracing of 3D neuron is essential for understanding the brain functions. While existing automatic tracing algorithms work well for the clean neuronal image with a single neuron, they are not robust to trace the neuron surrounded by nerve fibers. We propose a 3D U-Net-based network, namely 3D U-Net Plus, to segment the neuron from the surrounding fibers before the application of tracing algorithms. All the images in BigNeuron, the biggest available neuronal image dataset, contain clean neurons with no interference of nerve fibers, which are not practical to train the segmentation network. Based upon the BigNeuron images, we synthesize a SYNethic TAngled NEuronal Image dataset (SYNTANEI) to train the proposed network, by fusing the neurons with extracted nerve fibers. Due to the adoption of dropout, àtrous convolution and Àtrous Spatial Pyramid Pooling (ASPP), experimental results on the synthetic and real tangled neuronal images show that the proposed 3D U-Net Plus network achieved very promising segmentation results. The neurons reconstructed by the tracing algorithm using the segmentation result match significantly better with the ground truth than that using the original images.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2019.2926568DOI Listing
February 2020

Memristive Quantized Neural Networks: A Novel Approach to Accelerate Deep Learning On-Chip.

IEEE Trans Cybern 2021 Apr 17;51(4):1875-1887. Epub 2021 Mar 17.

Existing deep neural networks (DNNs) are computationally expensive and memory intensive, which hinder their further deployment in novel nanoscale devices and applications with lower memory resources or strict latency requirements. In this paper, a novel approach to accelerate on-chip learning systems using memristive quantized neural networks (M-QNNs) is presented. A real problem of multilevel memristive synaptic weights due to device-to-device (D2D) and cycle-to-cycle (C2C) variations is considered. Different levels of Gaussian noise are added to the memristive model during each adjustment. Another method of using memristors with binary states to build M-QNNs is presented, which suffers from fewer D2D and C2C variations compared with using multilevel memristors. Furthermore, methods of solving the sneak path issues in the memristive crossbar arrays are proposed. The M-QNN approach is evaluated on two image classification datasets, that is, ten-digit number and handwritten images of mixed National Institute of Standards and Technology (MNIST). In addition, input images with different levels of zero-mean Gaussian noise are tested to verify the robustness of the proposed method. Another highlight of the proposed method is that it can significantly reduce computational time and memory during the process of image recognition.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2019.2912205DOI Listing
April 2021

Asthmatic Augmentation of Airway Vagal Activity Involves Decreased Central Expression and Activity of CD73 in Rats.

ACS Chem Neurosci 2019 06 5;10(6):2809-2822. Epub 2019 Apr 5.

Department of Neurobiology, School of Basic Medical Sciences , Fudan University , Shanghai 200032 , China.

The severity of asthma is closely related to the intensity of airway vagal activity; however, it is unclear how airway vagal activity is centrally augmented in asthma. Here we report that in an asthma model of male Sprague-Dawley rats, the expression and activity of ecto-5'-nucleotidase (CD73) were decreased in airway vagal centers, ATP concentration in cerebral spinal fluid was increased, and the inhibitory and excitatory airway vagal responses to intracisternally injected ATP (5 μmol) and CD73 inhibitor AMPCP (5 μmol), respectively, were attenuated. In airway vagal preganglionic neurons (AVPNs) identified in medullary slices of neonatal Sprague-Dawley rats, AMPCP (100 μmol·L) caused excitatory effects, as are shown in patch-clamp by depolarization, increased neuronal discharge, and facilitated spontaneous excitatory postsynaptic currents (sEPSCs). In contrast, exogenous ATP (100 μmol·L, 1 mmol·L) primarily caused inhibitory effects, which are similar to those induced by exogenous adenosine (100 μmol·L). Adenosine A receptor antagonist CPT (5 μmol·L) blocked the inhibition of sEPSCs induced by 100 μmol·L exogenous ATP and that by 100 μmol·L exogenous adenosine, whereas 50 μmol·L CPT converted the inhibition of sEPSCs induced by 1 mmol·L ATP to facilitation that was blocked by addition of P2X receptor antagonist PPADS (20 μmol·L). These results demonstrate that in rat, the sEPSCs of AVPNs are facilitated by extracellular ATP via activation of P2X receptors and inhibited by extracellular adenosine via activation of A receptors; in experimental asthma, decreased CD73 expression and activity in airway vagal centers contribute to the augmentation of airway vagal activity through imbalanced ATP/ADO modulation of AVPNs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acschemneuro.9b00023DOI Listing
June 2019

Central blockade of the AT1 receptor attenuates pressor effects via reduction of glutamate release and downregulation of NMDA/AMPA receptors in the rostral ventrolateral medulla of rats with stress-induced hypertension.

Hypertens Res 2019 08 6;42(8):1142-1151. Epub 2019 Mar 6.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai, China.

Glutamatergic activity in the rostral ventrolateral medulla (RVLM), which is an important brain area where angiotensin II (Ang II) elicits its pressor effects, contributes to the onset of hypertension. The present study aimed to explore the effect of central Ang II type 1 receptor (ATR) blockade on glutamatergic actions in the RVLM of stress-induced hypertensive rats (SIHR). The stress-induced hypertension (SIH) model was established by electric foot shocks combined with noises. Normotensive Sprague-Dawley rats (control) and SIHR were intracerebroventricularly infused with the ATR antagonist candesartan or artificial cerebrospinal fluid for 14 days. Mean arterial pressure (MAP), heart rate (HR), plasma norepinephrine (NE), glutamate, and the expression of N-methyl-D-aspartic acid (NMDA) receptor subunit NR1, and α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptors in the RVLM increased in the SIH group. These increases were blunted by candesartan. Bilateral microinjection of the ionotropic glutamate receptor antagonist kynurenic acid, the NMDA receptor antagonist D-2-amino-5-phosphonopentanoate, or the AMPA/kainate receptors antagonist 6-cyano-7-nitroquinoxaline-2,3-dione into the RVLM caused a depressor response in the SIH group, but not in other groups. NR1 and AMPA receptors expressed in the glutamatergic neurons of the RVLM, and glutamate levels, increased in the intermediolateral column of the spinal cord of SIHR. Central Ang II elicits release of glutamate, which binds to the enhanced ionotropic NMDA and AMPA receptors via ATR, resulting in activation of glutamatergic neurons in the RVLM, increasing sympathetic excitation in SIHR.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41440-019-0242-6DOI Listing
August 2019

Acid-sensing ion channel 1a is involved in ischaemia/reperfusion induced kidney injury by increasing renal epithelia cell apoptosis.

J Cell Mol Med 2019 05 22;23(5):3429-3440. Epub 2019 Feb 22.

Department of Nephrology, Zhongshan Hospital, Fudan University, Shanghai, China.

Acidic microenvironment is commonly observed in ischaemic tissue. In the kidney, extracellular pH dropped from 7.4 to 6.5 within 10 minutes initiation of ischaemia. Acid-sensing ion channels (ASICs) can be activated by pH drops from 7.4 to 7.0 or lower and permeates to Ca entrance. Thus, activation of ASIC1a can mediate the intracellular Ca accumulation and play crucial roles in apoptosis of cells. However, the role of ASICs in renal ischaemic injury is unclear. The aim of the present study was to test the hypothesis that ischaemia increases renal epithelia cell apoptosis through ASIC1a-mediated calcium entry. The results show that ASIC1a distributed in the proximal tubule with higher level in the renal tubule ischaemic injury both in vivo and in vitro. In vivo, Injection of ASIC1a inhibitor PcTx-1 previous to ischaemia/reperfusion (I/R) operation attenuated renal ischaemic injury. In vitro, HK-2 cells were pre-treated with PcTx-1 before hypoxia, the intracellular concentration of Ca , mitochondrial transmembrane potential (∆ψm) and apoptosis was measured. Blocking ASIC1a attenuated I/R induced Ca overflow, loss of ∆ψm and apoptosis in HK-2 cells. The results revealed that ASIC1a localized in the proximal tubular and contributed to I/R induced kidney injury. Consequently, targeting the ASIC1a may prove to be a novel strategy for AKI patients.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/jcmm.14238DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6484315PMC
May 2019

Upregulation of AT Receptor Mediates a Pressor Effect Through ROS-SAPK/JNK Signaling in Glutamatergic Neurons of Rostral Ventrolateral Medulla in Rats With Stress-Induced Hypertension.

Front Physiol 2018 8;9:1860. Epub 2019 Jan 8.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai, China.

The present study examined whether angiotensin II (Ang II) mediates the pressor effect through nicotinamide adenine dinucleotide phosphate (NADPH) oxidase-derived reactive oxygen species (ROS)-mitogen-activated protein kinase (MAPK) signaling in the glutamatergic neurons of the rostral ventrolateral medulla (RVLM) in stress-induced hypertensive rats (SIHR). The SIHR model was established using electric foot-shocks combined with noises for 15 days. We observed that Ang II type 1 receptor (ATR) and the glutamatergic neurons co-localized in the RVLM of SIHR. Furthermore, glutamate levels in the intermediolateral column of the spinal cord were higher in SIHR than in controls. Microinjection of Ang II into the RVLM of SIHR activated stress-activated protein kinase/Jun N-terminal kinase (SAPK/JNK), extracellular signal-regulated protein kinase (ERK) 1/2, and p38MAPK. Compared with controls, the activation of SAPK/JNK, ERK1/2, p38MAPK, and ROS in the RVLM were higher in SIHR, an effect that was blocked by an NADPH oxidase inhibitor (apocynin) and an ATR antagonist (candesartan). RVLM microinjection of apocynin or a SAPK/JNK inhibitor (SP600125), but not an ERK1/2 inhibitor (U0126) or a p38MAPK inhibitor (SB203580), decreased ATR mRNA and mean arterial blood pressure (MABP) in SIHR. The increase of ATR protein expression and MABP was inhibited by intracerebroventricular infusion (ICV), for 14 days, of SP600125, but not U0126 or SB203580 in SIHR. We conclude that Ang II modulates the pressor effect through ATR-dependent ROS-SAPK/JNK signaling in glutamatergic neurons in the RVLM of SIHR.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fphys.2018.01860DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6331519PMC
January 2019

TASK1 and TASK3 Are Coexpressed With ASIC1 in the Ventrolateral Medulla and Contribute to Central Chemoreception in Rats.

Front Cell Neurosci 2018 29;12:285. Epub 2018 Aug 29.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai, China.

The ventrolateral medulla (VLM), including the lateral paragigantocellular nucleus (LPGi) and rostral VLM (RVLM), is commonly considered to be a chemosensitive region. However, the specific mechanism of chemoreception in the VLM remains elusive. Acid-sensing ion channels (ASICs), a family of voltage-independent proton-gated cation channels, can be activated by an external pH decrease to cause Na entry and induce neuronal excitability. TWIK-related acid-sensitive potassium channels (TASKs) are members of another group of pH-sensitive channels; in contrast to AISICs, they can be stimulated by pH increases and are inhibited by pH decreases in the physiological range. Our previous study demonstrated that ASICs take part in chemoreception. The aims of this study are to explore whether TASKs participate in the acid sensitivity of neurons in the VLM, thereby cooperating with ASICs. Our research demonstrated that TASKs, including TASK1 and TASK3, are colocalized with ASIC1 in VLM neurons. Blocking TASKs by microinjection of the non-selective TASK antagonist bupivacaine (BUP), specific TASK1 antagonist anandamide (AEA) or specific TASK3 antagonist ruthenium red (RR) into the VLM increased the integrated phrenic nerve discharge (iPND), shortened the inspiratory time (Ti) and enhanced the respiratory drive (iPND/Ti). In addition, microinjection of artificial cerebrospinal fluid (ACSF) at a pH of 7.0 or 6.5 prolonged Ti, increased iPND and enhanced respiratory drive, which were inhibited by the ASIC antagonist amiloride (AMI). By contrast, microinjection of alkaline ACSF decreased iPND and respiratory drive, which were inhibited by AEA. Taken together, our data suggest that TASK1 and TASK3 are coexpressed with ASIC1 in the VLM. Moreover, TASK1 and TASK3 contribute to the central regulation of breathing by coordinating with each other to perceive local pH changes; these results indicate a novel chemosensitive mechanism of the VLM.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fncel.2018.00285DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6123564PMC
August 2018

Skin Lesion Analysis towards Melanoma Detection Using Deep Learning Network.

Sensors (Basel) 2018 Feb 11;18(2). Epub 2018 Feb 11.

Computer Vision Institute, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China.

Skin lesions are a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging due to the following reasons: low contrast between lesions and skin, visual similarity between melanoma and non-melanoma lesions, etc. Hence, reliable automatic detection of skin tumors is very useful to increase the accuracy and efficiency of pathologists. In this paper, we proposed two deep learning methods to address three main tasks emerging in the area of skin lesion image processing, i.e., lesion segmentation (task 1), lesion dermoscopic feature extraction (task 2) and lesion classification (task 3). A deep learning framework consisting of two fully convolutional residual networks (FCRN) is proposed to simultaneously produce the segmentation result and the coarse classification result. A lesion index calculation unit (LICU) is developed to refine the coarse classification results by calculating the distance heat-map. A straight-forward CNN is proposed for the dermoscopic feature extraction task. The proposed deep learning frameworks were evaluated on the ISIC 2017 dataset. Experimental results show the promising accuracies of our frameworks, i.e., 0.753 for task 1, 0.848 for task 2 and 0.912 for task 3 were achieved.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/s18020556DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5855504PMC
February 2018

Emodin Inhibits ATP-Induced Proliferation and Migration by Suppressing P2Y Receptors in Human Lung Adenocarcinoma Cells.

Cell Physiol Biochem 2017 29;44(4):1337-1351. Epub 2017 Nov 29.

Department of Physiology and Pathophysiology, School of Basic Medical Sciences, Fudan University, Shanghai, China.

Background/aims: Extracellular ATP performs multiple important functions via activation of P2 receptors on the cell surface. P2Y receptors play critical roles in ATP evoked response in human lung adenocarcinoma cells (A549 cells). Emodin is an anthraquinone derivative originally isolated from Chinese rhubarb, possesses anticancer properties. In this study we examined the inhibiting effects of emodin on proliferation, migration and epithelial-mesenchymal transition (EMT) by suppressing P2Y receptors-dependent Ca2+ increase and nuclear factor-κB (NF-KB) signaling in A549 cells.

Methods: A549 cells were pretreated with emodin before stimulation with ATP for the indicated time. Then, intracellular Ca2+ concentration ([Ca2+]i) was measured by Fluo-8/AM staining. Cell proliferation and cell cycle progression were tested by CCK8 assay and flow cytometry In addition, wound healing and western blot were performed to determine cell migration and related protein levels (Bcl-2, Bax, claudin-1, NF-κB).

Results: Emodin blunted ATP/UTP-induced increase of [Ca2+]i and cell proliferation concentration-dependently Meanwhile, it decreased ATP-induced cells accumulation in the S phase. Furthermore, emodin altered protein abundance of Bcl-2, Bax and claudin-1 and attenuated EMT caused by ATP. Such ATP-induced cellular reactions were also inhibited by a nonselective P2Y receptors antagonist, suramin, in a similar way to emodin. Besides, emodin could inhibit activation of NF-κB, thus suppressed ATP-induced proliferation, migration and EMT.

Conclusion: Our results demonstrated that emodin inhibits ATP-induced proliferation, migration, EMT by suppressing P2Y receptors-mediated [Ca2+]i increase and NF-κB signaling in A549 cells.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1159/000485495DOI Listing
January 2018

Patchouli oil ameliorates acute colitis: A targeted metabolite analysis of 2,4,6-trinitrobenzenesulfonic acid-induced rats.

Exp Ther Med 2017 Aug 12;14(2):1184-1192. Epub 2017 Jun 12.

School of Chinese Materia Medica, Guangzhou University of Chinese Medicine, Guangzhou, Guangdong 510006, P.R. China.

The incidence of inflammatory bowel disease (IBD), characterized by chronic, relapsing intestinal inflammation, has continually increased in recent years. A previous study by our group identified five potential metabolic markers possibly associated with the pathology of 2,4,6-trinitrobenzenesulfonic acid (TNBS)-induced IBD in rats. The present study aimed to examine the potential therapeutic effects of the essential oil of (also known as patchouli; PO) on TNBS-induced rats and investigate the concomitant metabolic changes by targeting the previously identified potential markers. is widely used to treat gastrointestinal diseases, including IBD, in China. The results of the present study showed that PO (270 mg/kg, rectal instillation) significantly alleviated colonic damage and reduced disease activity indicators and colonic myeloperoxidase in TNBS-induced rats. In addition, a targeted metabolic profiling study identified that four metabolites were elevated in the urine of the animals in the TNBS group, which were significantly inhibited by treatment with PO: Two tryptophan metabolites [4-(2-aminophenyl)-2,4-dioxobutanoic acid and 4,6-cihydroxyquinoline] and two gut microbial metabolites (phenylacetylglycine and p-cresol glucuronide). Taken together, these findings suggested that PO ameliorated the symptoms of TNBS-induced IBD and reversed the metabolic changes potentially associated with TNBS-induced IBD in rats.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3892/etm.2017.4577DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5525581PMC
August 2017

Case study of 3D fingerprints applications.

PLoS One 2017 11;12(4):e0175261. Epub 2017 Apr 11.

Computer Vision Institute, School of Computer Science & Software Engineering, Shenzhen University, Shen Zhen, Guang Dong, China.

Human fingers are 3D objects. More information will be provided if three dimensional (3D) fingerprints are available compared with two dimensional (2D) fingerprints. Thus, this paper firstly collected 3D finger point cloud data by Structured-light Illumination method. Additional features from 3D fingerprint images are then studied and extracted. The applications of these features are finally discussed. A series of experiments are conducted to demonstrate the helpfulness of 3D information to fingerprint recognition. Results show that a quick alignment can be easily implemented under the guidance of 3D finger shape feature even though this feature does not work for fingerprint recognition directly. The newly defined distinctive 3D shape ridge feature can be used for personal authentication with Equal Error Rate (EER) of ~8.3%. Also, it is helpful to remove false core point. Furthermore, a promising of EER ~1.3% is realized by combining this feature with 2D features for fingerprint recognition which indicates the prospect of 3D fingerprint recognition.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0175261PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5388323PMC
September 2017

A 3-D Gabor Phase-Based Coding and Matching Framework for Hyperspectral Imagery Classification.

IEEE Trans Cybern 2018 Apr 28;48(4):1176-1188. Epub 2017 Mar 28.

As manual labeling is very difficult and time-consuming, the labeled samples used to train a supervised classifier are generally limited, which become one of the biggest challenge for hyperspectral imagery classification. In order to tackle this issue, a recent trend is to exploit the structure information of materials, as which reflects the region homogeneity in the spatial domain and offers an invaluable complement to the spectral information. In this respect, 3-D Gabor wavelets have been introduced to extract joint spectral-spatial features for hyperspectral images. One the one hand, the features extracted by 3-D Gabor wavelets lead to very good performance for classification. On the other hand, its drawbacks, i.e., big number of features and high computational cost, limit its applicability. In this paper, a 3-D Gabor-wavelet-based phase coding and Hamming distance-based matching (3DGPC-HDM) framework is developed for hyperspectral imagery classification. The proposed method, instead of taking into account the large volume of Gabor magnitude features, exploits the Gabor phase features with certain orientations (i.e., the direction parallel to the spectral axis), which are then encoded by a simple quadrant bit coding scheme. After that, a normalized Hamming distance matching (HDM) method is adopted to determine the similarity of two samples, and the nearest neighbor classifier is routinely utilized for pixelwise recognition. Finally, experiments on three real hyperspectral data sets show that the proposed 3DGPC-HDM leads to very good performance. Comparisons with the state-of-the-art methods in the literature, in terms of both classifier complexity and generalization ability from very small training sets, are also included.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TCYB.2017.2682846DOI Listing
April 2018
-->