Search

The Kullback-Leibler divergence (KLD), which is widely used to measure the similarity between two distributions, plays an important role in many applications. In this article, we address the KLD metric-learning task, which aims at learning the best KLD-type metric from the distributions of datasets. Concretely, first, we extend the conventional KLD by introducing a linear mapping and obtain the best KLD to well express the similarity of data distributions by optimizing such a linear mapping. It improves the expressivity of data distribution, which means it makes the distributions in the same class close and those in different classes far away. Then, the KLD metric learning is modeled by a minimization problem on the manifold of all positive-definite matrices. To deal with this optimization task, we develop an intrinsic steepest descent method, which preserves the manifold structure of the metric in the iteration. Finally, we apply the proposed method along with ten popular metric-learning approaches on the tasks of 3-D object classification and document classification. The experimental results illustrate that our proposed method outperforms all other methods.

In this paper, we propose a novel probabilistic variant of iterative closest point (ICP) dubbed as CoBigICP. The method leverages both local geometrical information and global noise characteristics. Locally, the 3D structure of both target and source clouds are incorporated into the objective function through bidirectional correspondence. Globally, error metric of correntropy is introduced as noise model to resist outliers. Importantly, the close resemblance between normal-distributions transform (NDT) and correntropy is revealed. To ease the minimization step, an on-manifold parameterization of the special Euclidean group is proposed. Extensive experiments validate that CoBigICP outperforms several well-known and state-of-the-art methods., Comment: 6 pages, 4 figures. Accepted to IROS2020

Single-modal neuroimaging-based diagnosis for brain diseases is a main routine due to the lack of advanced imaging devices, especially in rural hospitals. Transfer learning (TL) has demonstrate edits effectiveness in improving the performance of a computer-aided diagnosis (CAD) system, which transfers knowledge from the model of a related imaging modality (source domain (SD)) to that of the diagnosis modality (target domain (TD)). Multiple empirical kernel learning machine (MEKLM) is a newly proposed classifier with superior performance to the conventional classifiers with implicit kernel mapping. In this work, we propose a novel projective model(PM) based sparse MEKLM(PM-SMEKLM) algorithm to learn a cross-domain transformation by PM in way of the parameter-based TL, and then apply it to the neuroimaging-based CAD for brain diseases. Sparse learning is integrated into MEKLM to further select effective information in SD for knowledge transfer and enhance the generalization ability of the classifier in TD. The projection matrix in PM and sparse representations in MEKLM are jointly learned, which effectively improves the performance of PM-SMEKLM. Three experiments are conducted on two neuroimaging datasets for the diagnosis of Alzheimer’s disease and Parkinson’s disease, respectively. Experimental results show that the proposed PM-SMEKLM algorithm outperforms all the compared algorithms.

Elastography ultrasound (EUS) imaging has shown its effectiveness for diagnosis of tumors by providing additional information about tissue stiffness to the conventional B-mode ultrasound (BUS). However, due to the lack of EUS devices and experienced sonologists, EUS is not widely used, especially in rural areas. It is still a challenging task to improve the performance of the single-modal BUS-based computer-aided diagnosis (CAD) for tumors. In this work, we propose a novel transfer learning (TL)–based deep neural network (DNN) algorithm, named CW-PM-DNN, for the BUS-based CAD by transferring diagnosis knowledge from EUS during model training. CW-PM-DNN integrates both the feature-level and classifier-level knowledge transfer into a unified framework. In the feature-level TL, a bichannel DNN is learned by the cross-weight-based multimodal DL (MDL-CW) algorithm to transfer informative features from EUS to BUS. In the classifier-level TL, a projective model (PM)–based classifier is then embedded to the pretrained bichannel DNN to implement the parameter transfer in the classifier model at the second stage. The back-propagation procedure is then applied to optimize the whole CW-PM-DNN to further improve its performance. Experimental results on two bimodal ultrasound tumor datasets demonstrate that the proposed CW-PM-DNN achieves the best classification accuracy, sensitivity, and specificity of 89.02 ± 1.54%, 88.37 ± 4.72%, and 89.63 ± 4.06%, respectively, for the breast ultrasound dataset, and the corresponding values of 80.57 ± 3.41%, 76.67 ± 3.85%, and 83.94 ± 3.95%, respectively, for the prostate ultrasound dataset. The proposed two-stage TL-based CW-PM-DNN algorithm outperforms all the compared algorithms. It is also proved that the performance of the BUS-based CAD can be significantly improved by transferring the knowledge of EUS. It suggests that CW-PM-DNN has the potential for more applications in the field of medical image–based CAD.

In this paper, we introduce a modified Generalized Iterative Closest Point (GICP) algorithm by presenting a coarse-to-fine strategy. Our contributions can be summarized as: Firstly, we use adaptively a plane-to-plane probabilistic matching model by gradually reducing the neighborhood range for given two point sets. It is an inner coarse-to-fine iteration process. Secondly, we use an outer coarse-to-fine strategy to bridge the point-to-point and plane-to-plane registration for refining the matching. Thirdly, we use the trimmed method to gradually eliminate the effects of incorrect correspondences, which improves the robustness of the methods especially for the low overlap cases. Moreover, we also extend our method to the scale registration case. Finally, we conduct extensive experiments to demonstrate that our method is more reliable and robust in various situations, including missing points, noise and different scale factors. Experimental results show that our approach outperforms several state-of-the-art registration methods.

Partial registration for point clouds plays an important role in various fields such as 3D mapping reconstruction, remote sensing, unmanned driving, and cultural heritage protection. Unfortunately, partial registration is challenging due to difficulties such as the low overlap ratio of two point clouds and the perturbation in the orderless and sparse 3D point clouds. Thus, a variety of the 3D shape context descriptors are introduced for finding the optimal matching. However, extracting geometric features and descriptors are time consuming and easily degenerated by noise. To overcome these problems, we introduce a parallel coarse-to-fine partial registration method. Our contributions can be summarized as: Firstly, a robust coarse trimmed method is proposed to estimate the coarse overlap area and the initial transformation via fast bilateral denoising and parallel point feature histogram (PPFH) descriptor aligning. Secondly, an accelerated fine registration procedure is conducted by a parallel trimmed iterative closest point (PTrICP) method. Moreover, most parts of our coarse-to-fine workflow are accelerated under the Graphics Processing Unit (GPU) parallel execution mode for efficiency. Thirdly, we extend our method from the rigid registration to the isotropic scaling registration, which improves its applicability. Experiments have demonstrated that our method is feasible and robust in various situations, including the low overlap ratio, outlier, noise and scaling.

Transcranial sonography (TCS) is a valid neuroimaging tool for the diagnosis of Parkinson’s disease (PD). The TCS-based computer-aided diagnosis (CAD) has attracted increasing attention in recent years, in which feature representation and pattern classification are two critical issues. Deep polynomial network (DPN) is a newly proposed deep learning algorithm that has shown its advantage in learning effective feature representation for samples with a small size. In this work, an improved DPN algorithm with enhanced performance on both feature representation and classification is proposed. First, the empirical kernel mapping (EKM) algorithm is embedded into DPN (EKM-DPN) to improve its feature representation. Second, the network pruning strategy is utilized in the EKM-DPN (named P-EKM-DPN). It not only produces robust feature representation, but also addresses the overfitting issues for the subsequent classifiers to some extent. Lastly, the generalization ability is further enhanced by applying the Dropout approach to P-EKM-DPN (D-P-EKM-DPN). The proposed D-P-EKM-DPN algorithm has been evaluated on a TCS dataset with 153 samples. The experimental results indicate that D-P-EKM-DPN outperforms all the compared algorithms and achieves the best classification accuracy, sensitivity, and specificity of 86.95 ± 3.15%, 85.77 ± 7.87%, and 87.16 ± 6.50%, respectively. The proposed D-P-EKN-DPN algorithm has a great potential in TCS-based CAD for PD due to its excellent performance.

Purpose Partial alignment for 3 D point sets is a challenging problem for laser calibration and robot calibration due to the unbalance of data sets, especially when the overlap of data sets is low. Geometric features can promote the accuracy of alignment. However, the corresponding feature extraction methods are time consuming. The purpose of this paper is to find a framework for partial alignment by an adaptive trimmed strategy. Design/methodology/approach First, the authors propose an adaptive trimmed strategy based on point feature histograms (PFH) coding. Second, they obtain an initial transformation based on this partition, which improves the accuracy of the normal direction weighted trimmed iterative closest point (ICP) method. Third, they conduct a series of GPU parallel implementations for time efficiency. Findings The initial partition based on PFH feature improves the accuracy of the partial registration significantly. Moreover, the parallel GPU algorithms accelerate the alignment process. Research limitations/implications This study is applicable to rigid transformation so far. It could be extended to non-rigid transformation. Practical implications In practice, point set alignment for calibration is a technique widely used in the fields of aircraft assembly, industry examination, simultaneous localization and mapping and surgery navigation. Social implications Point set calibration is a building block in the field of intelligent manufacturing. Originality/value The contributions are as follows: first, the authors introduce a novel coarse alignment as an initial calibration by PFH descriptor similarity, which can be viewed as a coarse trimmed process by partitioning the data to the almost overlap part and the rest part; second, they reduce the computation time by GPU parallel coding during the acquisition of feature descriptor; finally, they use the weighted trimmed ICP method to refine the transformation.

Longitudinal image analysis plays an important role in depicting the development of the brain structure, where image regression and interpolation are two commonly used techniques. In this paper, we develop an efficient model and approach based on a path regression on the image manifold instead of the geodesic regression to avoid the complexity of the geodesic computation. Concretely, first we model the deformation by diffeomorphism; then, a large deformation is represented by a path on the orbit of the diffeomorphism group action. This path is obtained by compositing several small deformations, which can be well approximated by its linearization. Second, we introduce some intermediate images as constraints to the model, which guides to form the best-fitting path. Thirdly, we propose an approximated quadratic model by local linearization method, where a closed form is deduced for the solution. It actually speeds up the algorithm. Finally, we evaluate the proposed model and algorithm on a synthetic data and a real longitudinal MRI data. The results show that our proposed method outperforms several state-of-the-art methods.

Histopathological image analysis works as ‘gold standard’ for cancer diagnosis. Its computer-aided approach has attracted considerable attention in the field of digital pathology, which highly depends on the feature representation for histopathological images. The principal component analysis network (PCANet) is a novel unsupervised deep learning framework that has shown its effectiveness for feature representation learning. However, PCA is susceptible to noise and outliers to affect the performance of PCANet. The Grassmann average (GA) is superior to PCA on robustness. In this work, a GA network (GANet) algorithm is proposed by embedding GA algorithm into the PCANet framework. Moreover, since quaternion algebra is an excellent tool to represent color images, a quaternion-based GANet (QGANet) algorithm is further developed to learn effective feature representations containing color information for histopathological images. The experimental results based on three histopathological image datasets indicate that the proposed QGANet achieves the best performance on the classification of color histopathological images among all the compared algorithms.

The accuracy of classification and retrieval significantly depends on the metric used to compute the similarity between samples. For preserving the geometric structure, the symmetric positive definite (SPD) manifold is introduced into the metric learning problem. However, the SPD constraint is too strict to describe the real data distribution. In this paper, we extend the intrinsic metric learning problem to semi-definite case, by which the data distribution is better described for various classification tasks. First, we formulate the metric learning as a minimization problem to the SPD manifold on subspace, which not only considers to balance the information between inner classes and inter classes by an adaptive tradeoff parameter but also improves the robustness by the low-rank subspaces presentation. Thus, it benefits to design a structure-preserving algorithm on subspace by using the geodesic structure of the SPD subspace. To solve this model, we develop an iterative strategy to update the intrinsic metric and the subspace structure, respectively. Finally, we compare our proposed method with ten state-of-the-art methods on four data sets. The numerical results validate that our method can significantly improve the description of the data distribution, and hence, the performance of the image classification task.

Neuroimaging has shown its effectiveness for diagnosis of Parkinson's disease (PD), and the neuroimaging-based computer-aided diagnosis (CAD) then attracts considerable attention. In a CAD system, the classifier module is one of the key components, which directly decides the classification performance. As a newly proposed classifier, the large margin distribution machine (LDM) has excellent generalization by maximizing the margin mean and minimizing the margin variance simultaneously. However, LDM still suffers from the problem of kernel selection. In this work, we propose a deep neural mapping large margin distribution machine (DNMLDM) algorithm by adopting the deep neural network (DNN) to perform a kernel mapping instead of the implicit kernel function in LDM. A two-stage joint training strategy is then developed, including the unsupervised layer-wise pre-training for DNN and then the supervised fine-tuning for all parameters in the whole networks. Two real-world PD datasets, namely the transcranial sonography (TCS) dataset and the magnetic resonance imaging (MRI) dataset, are used to evaluate the performance of DNMLDM algorithm. The experimental results show that the proposed DNMLDM outperforms all the compared algorithms on both datasets.

B-mode ultrasound (BUS) imaging is a routine tool for diagnosis of liver cancers, while contrast-enhanced ultrasound (CEUS) provides additional information to BUS on the local tissue vascularization and perfusion to promote diagnostic accuracy. In this work, we propose to improve the BUS-based computer aided diagnosis for liver cancers by transferring knowledge from the multi-view CEUS images, including the arterial phase, portal venous phase, and delayed phase, respectively. To make full use of the shared labels of paired of BUS and CEUS images to guide knowledge transfer, support vector machine plus (SVM+), a specifically designed transfer learning (TL) classifier for paired data with shared labels, is adopted for this supervised TL. A nonparallel hyperplane based SVM+ (NHSVM+) is first proposed to improve the TL performance by transferring the per-class knowledge from source domain to the corresponding target domain. Moreover, to handle the issue of multi-source TL, a multi-kernel learning based NHSVM+ (MKL-NHSVM+) algorithm is further developed to effectively transfer multi-source knowledge from multi-view CEUS images. The experimental results indicate that the proposed MKL-NHSVM+ outperforms all the compared algorithms for diagnosis of liver cancers, whose mean classification accuracy, sensitivity, and specificity are 88.18 ± 3.16 %, 86.98 ± 4.77 %, and 89.42±3.77%, respectively.

The behavioral and cognitive deficits in autism spectrum disorder (ASD) patients are associated with abnormal brain function. The resting-state functional magnetic resonance imaging (rs-fMRI) is an effective non-invasive tool for revealing the brain dysfunction for ASD patients. However, most rs-fMRI based ASD diagnosis methods are developed for simple binary classification, instead of classification of multiple sub-types in ASD. Besides, they assume that the class boundary in ASD classification is crisp, whereas the symptoms of ASD sub-types are a continuum from mild to severe impairments in both social communication and restrictive repetitive behaviors/interests, and do not have crisp boundary between each other. To this end, we introduce label distribution learning (LDL) into multi-class ASD classification and propose LDL-CSCS under the LDL framework. Specifically, the label distribution is introduced to describe how individual disease labels correlate with the subject. In the learning crierion of LDL-CSCS, the label distribution is decomposed into the class-shared and class-specific components, in which the class-shared component records the common knowledge across all persons and the class-specific component records the specific information in each ASD sub-type. Low-rank constraint is imposed on the class-shared component whereas the group sparse constraint is imposed on the class-specific component, respectively. An Augmented Lagrange Method (ALM) is developed to find the optimal solution. The experimental results show that the proposed method for ASD diagnosis has superior classification performance, compared with some existing algorithms.

Existing methods on facial expression recognition (FER) are mainly trained in the setting when all expression classes are fixed in advance. However, in real applications, expression classes are becoming increasingly fine-grained and incremental. To deal with sequential expression classes, we can fine-tune or re-train these models, but this often results in poor performance or large computing resources consumption. To address these problems, we develop an Incremental Facial Expression Recognition Network (IExpressNet), which can learn a competitive multi-class classifier at any time with a lower requirement of computing resources. Specifically, IExpressNet consists of two novel components. First, we construct an exemplar set by dynamically selecting representative samples from old expression classes. Then, the exemplar set and new expression classes samples constitute the training set. Second, we design a novel center-expression-distilled loss. As for facial expression in the wild, center-expression-distilled loss enhances the discriminative power of the deeply learned features and prevents catastrophic forgetting. Extensive experiments are conducted on two large-scale FER datasets in the wild, RAF-DB and AffectNet. The results demonstrate the superiority of the proposed method as compared to state-of-the-art incremental learning approaches.

Colorectal cancer is the second and the third most common cancer in women and men, respectively. Pathological diagnosis is the “gold standard” for tumor diagnosis. Accurate segmentation of glands from tissue images is a crucial step in assisting pathologists in their diagnosis. The typical methods for gland segmentation form a dense image representation, ignoring its texture and multi-scale attention information. Therefore, we utilize a Gabor-based module to extract texture information at different scales and directions in histopathology images. This paper also designs a Cascade Squeeze Bi-Attention (CSBA) module. Specifically, we add Atrous Cascade Spatial Pyramid (ACSP), Squeeze Position Attention (SPA) module and Squeeze Channel Attention module (SCA) to model semantic correlation and maintain the multi-level aggregation on the spatial pyramid with different dilations. Besides, to solve the imbalance of data distribution and boundary blur, we propose a hybrid loss function to response the object boudary better. The experimental results show that the proposed method achieves state-of-the-art performance on the GlaS challenge dataset and CRAG colorectal adenocarcinoma dataset, respectively.

In the field of autonomous driving, carriers are equipped with a variety of sensors, including cameras and LiDARs. However, the camera suffers from problems of illumination and occlusion, and the LiDAR encounters motion distortion, degenerate environment and limited ranging distance. Therefore, fusing the information from these two sensors deserves to be explored. In this paper, we propose a fusion network which robustly captures both the image and point cloud descriptors to solve the place recognition problem. Our contribution can be summarized as: (1) applying the trimmed strategy in the point cloud global feature aggregation to improve the recognition performance, (2) building a compact fusion framework which captures both the robust representation of the image and 3D point cloud, and (3) learning a proper metric to describe the similarity of our fused global feature. The experiments on KITTI and KAIST datasets show that the proposed fused descriptor is more robust and discriminative than the single sensor descriptor.

Elastography ultrasound (EUS) provides additional bio-mechanical information about lesion for B-mode ultrasound (BUS) in the diagnosis of breast cancers. However, joint utilization of both BUS and EUS is not popular due to the lack of EUS devices in rural hospitals, which arouses a novel modality imbalance problem in computer-aided diagnosis (CAD) for breast cancers. Current transfer learning (TL) pay little attention to this special issue of clinical modality imbalance, that is, the source domain (EUS modality) has fewer labeled samples than those in the target domain (BUS modality). Moreover, these TL methods cannot fully use the label information to explore the intrinsic relation between two modalities and then guide the promoted knowledge transfer. To this end, we propose a novel doubly supervised TL network (DDSTN) that integrates the Learning Using Privileged Information (LUPI) paradigm and the Maximum Mean Discrepancy (MMD) criterion into a unified deep TL framework. The proposed algorithm can not only make full use of the shared labels to effectively guide knowledge transfer by LUPI paradigm, but also perform additional supervised transfer between unpaired data. We further introduce the MMD criterion to enhance the knowledge transfer. The experimental results on the breast ultrasound dataset indicate that the proposed DDSTN outperforms all the compared state-of-the-art algorithms for the BUS-based CAD.

Metric-based few-shot methods learn to recognize object categories from one or a few examples according to the distances between class features and query samples. The predicting labels of query samples are the same as those of the nearest class features. However, a single metric criterion can not model the distributions of different datasets very well. To get the more suitable metrics for a specified dataset, we propose a Multi-Metric Joint Discrimination Network (MMJDN) in this paper. Firstly, Deep Metric Module (DMM) is introduced to catch the complex relation between each pair of class features and query samples. Secondly, Adaptive Weights Module (AWM) is proposed to generate adaptive weights for different metric criteria. Our method is evaluated on three datasets: miniImageNet, Fewshot-Cifar100 (FC100) and Virus Texture Dataset (Virus15). The experimental results show that MMJDN provides positive performance for few-shot learning compared with some baselines.

The accurate diagnosis of Alzheimer's disease (AD) and its early stage, i.e., mild cognitive impairment, is essential for timely treatment and possible delay of AD. Fusion of multimodal neuroimaging data, such as magnetic resonance imaging (MRI) and positron emission tomography (PET), has shown its effectiveness for AD diagnosis. The deep polynomial networks (DPN) is a recently proposed deep learning algorithm, which performs well on both large-scale and small-size datasets. In this study, a multimodal stacked DPN (MM-SDPN) algorithm, which MM-SDPN consists of two-stage SDPNs, is proposed to fuse and learn feature representation from multimodal neuroimaging data for AD diagnosis. Specifically speaking, two SDPNs are first used to learn high-level features of MRI and PET, respectively, which are then fed to another SDPN to fuse multimodal neuroimaging information. The proposed MM-SDPN algorithm is applied to the ADNI dataset to conduct both binary classification and multiclass classification tasks. Experimental results indicate that MM-SDPN is superior over the state-of-the-art multimodal feature-learning-based algorithms for AD diagnosis.

The bimodal ultrasound, namely B-mode ultrasound (BUS) and elastography ultrasound (EUS), provide complementary information to improve the diagnostic accuracy of breast cancers. However, in clinical practice, it is easier to acquire the labeled BUS images than the paired bimodal ultrasound data with shared labels due to the lack of EUS devices, especially in many rural hospitals. Thus, the single-modal BUS-based computer-aided diagnosis (CAD) generally has wide applications. Transfer learning (TL) can promote a BUS-based CAD model by transferring additional knowledge from EUS modality. To make full use of labeled paired bimodal data and the additional single-modal BUS images for knowledge transfer, a novel doubly supervised parameter transfer classifier (DSPTC) is proposed to well handle the TL between imbalanced modalities with the guidance of label information. Specifically, the proposed DSPTC consists of two loss functions corresponding to the paired bimodal ultrasound data with shared labels and the unpaired images with different labels, respectively. The former uses the loss function in the specially designed TL paradigm of support vector machine plus, while the latter adopts the Hilbert-Schmidt Independence Criterion (HSIC) for knowledge transfer between the unpaired images, which consist of the single-modal BUS images and the EUS images from the paired bimodal data. Consequently, the doubly supervised knowledge transfer is implemented by way of parameter transfer in a unified optimization framework. Two experiments are designed to evaluate the proposed DSPTC for the ultrasound-based diagnosis of breast cancers. The experimental results indicate that DSPTC outperforms all the compared algorithms, suggesting its wide potential applications.

Deep learning has been successfully applied to the single-image super-resolution (SISR) task with superior performance in recent years. However, most convolutional neural network (CNN) based SR models have a large number of parameters to be optimized, which requires heavy computation and thereby limits their real-world applications. In this work, a novel lightweight SR network, named Adaptive Weighted Super-Resolution Network (LW-AWSRN), is proposed to address this issue. A novel local fusion block (LFB) is developed in LW-AWSRN for efficient residual learning, which consists of several stacked adaptive weighted residual units (AWRU) and a local residual fusion unit (LRFU). Moreover, an adaptive weighted multi-scale (AWMS) module is proposed to make full use of features for the reconstruction of HR images. The AWMS module includes several convolutions with multiple scales, and the redundancy scale branch can be removed according to the contribution of adaptive weights for the lightweight network. The experimental results on the commonly used datasets show that the proposed LW-AWSRN achieves superior performance on × 2, × 3, × 4, and × 8 scale factors compared to state-of-the-art methods with similar parameters and computational overhead. It suggests that LW-AWSRN has a better trade-off between reconstruction quality and model size.

In this paper, we address the unsupervised subspace learning from a geometric viewpoint. First, we formulate the subspace learning as an inverse problem on Grassmannian manifold by considering all subspaces as points on it. Then, to make the model computable, we parameterize the Grassmannian manifold by using an orbit of rotation group action on all standard subspaces, which are spanned by the orthonormal basis. Further, to improve the robustness, we introduce a low-rank regularizer which makes the dimension of subspace as low as possible. Thus, the subspace learning problem is transferred to a minimization problem with variables of rotation and dimension. Then, we adopt the alternately iterative strategy to optimize the variables, where a structure-preserving method, based on the geodesic structure of the rotation group, is designed to update the rotation. Finally, we compare the proposed approach with six state-of-the-art methods on three different kinds of real datasets. The experimental results validate that our proposed method outperforms all compared methods.

Unsupervised domain adaptation is a problem which exploits the knowledge learned from the resource-rich domain to obtain an accurate classifier for the resource-poor domain. Most of the existing methods lift performance by reducing the differences between distributions, such as the difference between marginal probability distributions, the difference between conditional probability distributions, or both. However, all these methods consider the two distributions to be equally important, which could lead to poor classification performance in practical applications. Therefore, a balanced factor is required to weigh the two distributions to compensate for the degraded performance. In this paper, we first introduce this balance factor to weigh the distribution importance. On this base, we utilize the marginal distribution, introduce the ideas of manifold regularization, and then preserve the neighboring structures of the data sets, with the dimension reduction as much as possible. By this way, we propose the manifold alignment and balanced distribution adaptation algorithm. A large number of experiments have also been conducted, showing that our algorithm behaves much better than the previous ones.

Spatial resolution is a critical imaging parameter in magnetic resonance imaging. The image super-resolution (SR) is an effective and cost efficient alternative technique to improve the spatial resolution of MR images. Over the past several years, the convolutional neural networks (CNN)-based SR methods have achieved state-of-the-art performance. However, CNNs with very deep network structures usually suffer from the problems of degradation and diminishing feature reuse, which add difficulty to network training and degenerate the transmission capability of details for SR. To address these problems, in this work, a progressive wide residual network with a fixed skip connection (named FSCWRN) based SR algorithm is proposed to reconstruct MR images, which combines the global residual learning and the shallow network based local residual learning. The strategy of progressive wide networks is adopted to replace deeper networks, which can partially relax the above-mentioned problems, while a fixed skip connection helps provide rich local details at high frequencies from a fixed shallow layer network to subsequent networks. The experimental results on one simulated MR image database and three real MR image databases show the effectiveness of the proposed FSCWRN SR algorithm, which achieves improved reconstruction performance compared with other algorithms.

Spatial resolution is one of the key parameters of magnetic resonance imaging (MRI). The image super-resolution (SR) technique offers an alternative approach to improve the spatial resolution of MRI due to its simplicity. Convolutional neural networks (CNN)-based SR algorithms have achieved state-of-the-art performance, in which the global residual learning (GRL) strategy is now commonly used due to its effectiveness for learning image details for SR. However, the partial loss of image details usually happens in a very deep network due to the degradation problem. In this work, we propose a novel residual learning-based SR algorithm for MRI, which combines both multi-scale GRL and shallow network block-based local residual learning (LRL). The proposed LRL module works effectively in capturing high-frequency details by learning local residuals. One simulated MRI dataset and two real MRI datasets have been used to evaluate our algorithm. The experimental results show that the proposed SR algorithm achieves superior performance to all of the other compared CNN-based SR algorithms in this work.

The computer-aided quantitative analysis for histopathological images has attracted considerable attention. The stain decomposition on histopathological images is usually recommended to address the issue of co-localization or aliasing of tissue substances. Although the convolutional neural networks (CNN) is a popular deep learning algorithm for various tasks on histopathological image analysis, it is only directly performed on histopathological images without considering stain decomposition. The bilinear CNN (BCNN) is a new CNN model for fine-grained classification. BCNN consists of two CNNs, whose convolutional-layer outputs are multiplied with outer product at each spatial location. In this work, we propose a novel BCNN-based method for classification of histopathological images, which first decomposes histopathological images into hematoxylin and eosin stain components, and then perform BCNN on the decomposed images to fuse and improve the feature representation performance. The experimental results on the colorectal cancer histopathological image dataset with eight classes indicate that the proposed BCNN-based algorithm is superior to the traditional CNN.

In clinical practice, the magnetic resonance imaging (MRI) is a prevalent neuroimaging technique for Alzheimer's disease (AD) diagnosis. As a learning using privileged information (LUPI) algorithm, SVM+ has shown its effectiveness on the classification of brain disorders, with single-modal neuroimaging samples for testing but multimodal neuroimaging samples for training. In this work, we propose to apply the multimodal restricted Boltzmann machines (RBM) as an LUPI algorithm for feature learning so as to form an RBM+ algorithm. Furthermore, an ensemble LUPI algorithm is developed, integrating SVM+ and RBM+ by the multiple kernel boosting based strategy. The experimental results demonstrate that the proposed RBM+ works well as an LUPI algorithm for feature learning, and the ensemble LUPI algorithm is superior to the traditional predictive models for the MRI-based AD diagnosis using the positron emission tomography as the privileged information.

Nonlinear registration is an important technique to align two different images and widely applied in medical image analysis. In this paper, we develop a novel nonlinear registration framework based on the diffeomorphic demons, where a reciprocal regularizer is introduced to assume that the deformation between two images is an exact diffeomorphism. In detail, first, we adopt a bidirectional metric to improve the symmetry of the energy functional, whose variables are two reciprocal deformations. Secondly, we slack these two deformations into two independent variables and introduce a reciprocal regularizer to assure the deformations being the exact diffeomorphism. Then, we utilize an alternating iterative strategy to decouple the model into two minimizing subproblems, where a new closed form for the approximate velocity of deformation is calculated. Finally, we compare our proposed algorithm on two data sets of real brain MR images with two relative and conventional methods. The results validate that our proposed method improves accuracy and robustness of registration, as well as the gained bidirectional deformations are actually reciprocal.

With a rapid evolution of computer science, digital image manipulation becomes much more easily. One of the most commonly tampering with a digital image is copy-move forgery, in which a part of image is copied and used to cover another part in a different location in the same image. In this paper, we present an efficient and robust quaternion discrete cosine transform (QDCT) based method to detect this artifact. First, the R,G,B component of a tampered image is divided into overlapping fixed-size blocks, and a quaternion is constructed by using each R,G,B component block. Secondly, QDCT is applied to each quaternion block and using Zig-zag scanning extract finite QDCT coefficients, thus, the feature vectors are extracted from finite QDCT coefficients of overlapping blocks, and reduced features are generated. Finally, feature vectors are sorted lexicographically. The duplicated image blocks will be matched by a preset threshold value. Experiment results show that our proposed scheme can efficiently detect the copy-move forgery.

SUMMARY Weighted graph matching is computationally challenging due to the combinatorial nature of the set of permutations. In this paper, a new relaxation approach to weighted graph matching is proposed, by which a new matching algorithm, named alternate iteration algorithm, is designed. It is proved that the algorithm proposed is locally convergent. Experiments are presented to show the effectiveness of the proposed algorithm. eration algorithm.

Background Electroencephalography (EEG) based sleep staging is commonly used in clinical routine. Feature extraction and representation plays a crucial role in EEG-based automatic classification of sleep stages. Sparse representation (SR) is a state-of-the-art unsupervised feature learning method suitable for EEG feature representation. New method Collaborative representation (CR) is an effective data coding method used as a classifier. Here we use CR as a data representation method to learn features from the EEG signal. A joint collaboration model is established to develop a multi-view learning algorithm, and generate joint CR (JCR) codes to fuse and represent multi-channel EEG signals. A two-stage multi-view learning-based sleep staging framework is then constructed, in which JCR and joint sparse representation (JSR) algorithms first fuse and learning the feature representation from multi-channel EEG signals, respectively. Multi-view JCR and JSR features are then integrated and sleep stages recognized by a multiple kernel extreme learning machine (MK-ELM) algorithm with grid search. Results The proposed two-stage multi-view learning algorithm achieves superior performance for sleep staging. With a K-means clustering based dictionary, the mean classification accuracy, sensitivity and specificity are 81.10 ± 0.15%, 71.42 ± 0.66% and 94.57 ± 0.07%, respectively; while with the dictionary learned using the submodular optimization method, they are 80.29 ± 0.22%, 71.26 ± 0.78% and 94.38 ± 0.10%, respectively. Comparison with existing methods The two-stage multi-view learning based sleep staging framework outperforms all other classification methods compared in this work, while JCR is superior to JSR. Conclusions The proposed multi-view learning framework has the potential for sleep staging based on multi-channel or multi-modality polysomnography signals.

Effective and efficient spatial normalization of a large population of brain images is critical for many clinical and research studies, but it is technically very challenging. A commonly used approach is to choose a certain image as the template and then align all other images in the population to this template by applying pairwise registration. To avoid the potential bias induced by the inappropriate template selection, groupwise registration methods have been proposed to simultaneously register all images to a latent common space. However, current groupwise registration methods do not make full use of image distribution information for more accurate registration. In this paper, we present a novel groupwise registration method that harnesses the image distribution information by capturing the image distribution manifold using a hierarchical graph with its nodes representing the individual images. More specifically, a low-level graph describes the image distribution in each subgroup, and a high-level graph encodes the relationship between representative images of subgroups. Given the graph representation, we can register all images to the common space by dynamically shrinking the graph on the image manifold. The topology of the entire image distribution is always maintained during graph shrinkage. Evaluations on two datasets, one for 80 elderly individuals and one for 285 infants, indicate that our method can yield promising results.

The iterative closest point (ICP) algorithm is an accurate and fast approach for registration between two point sets in a same scale, but it doesn't handle the case with different scales. This paper instead introduces a novel approach named the iterative closest point with bounded scale (ICPBS) algorithm which integrates a scale with boundaries into the traditional ICP algorithm. This proposed technique uses the singular value decomposition algorithm and the properties of parabola to compute the similar transformation at each iterative step, and yields more satisfying robust results than the traditional ICP method in registration between two m-D point sets with different scales. Experimental results demonstrate the presented method is robust and fast for practical use.