Pairwise machine learning-based automatic diagnostic platform utilizing CT images and clinical information for predicting radiotherapy locoregional recurrence in elderly esophageal cancer patients.

OBJECTIVE: To investigate the feasibility and accuracy of predicting locoregional recurrence (LR) in elderly patients with esophageal squamous cell cancer (ESCC) who underwent radical radiotherapy using a pairwise machine learning algorithm. METHODS: The 130 datasets enrolled were randomly divided into a training set and a testing set in a 7:3 ratio. Clinical factors were included and radiomics features were extracted from pretreatment CT scans using pyradiomics-based software, and a pairwise naive Bayes (NB) model was developed. The performance of the model was evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA). To facilitate practical application, we attempted to construct an automated esophageal cancer diagnosis system based on trained models. RESULTS: To the follow-up date, 64 patients (49.23%) had experienced LR. Ten radiomics features and two clinical factors were selected for modeling. The model demonstrated good prediction performance, with area under the ROC curve of 0.903 (0.829-0.958) for the training cohort and 0.944 (0.849-1.000) for the testing cohort. The corresponding accuracies were 0.852 and 0.914, respectively. Calibration curves showed good agreement, and DCA curve confirmed the clinical validity of the model. The model accurately predicted LR in elderly patients, with a positive predictive value of 85.71% for the testing cohort. CONCLUSIONS: The pairwise NB model, based on pre-treatment enhanced chest CT-based radiomics and clinical factors, can accurately predict LR in elderly patients with ESCC. The esophageal cancer automated diagnostic system embedded with the pairwise NB model holds significant potential for application in clinical practice.

Structure Embedded Nucleus Classification for Histopathology Images.

Nuclei classification provides valuable information for histopathology image analysis. However, the large variations in the appearance of different nuclei types cause difficulties in identifying nuclei. Most neural network based methods are affected by the local receptive field of convolutions, and pay less attention to the spatial distribution of nuclei or the irregular contour shape of a nucleus. In this paper, we first propose a novel polygon-structure feature learning mechanism that transforms a nucleus contour into a sequence of points sampled in order, and employ a recurrent neural network that aggregates the sequential change in distance between key points to obtain learnable shape features. Next, we convert a histopathology image into a graph structure with nuclei as nodes, and build a graph neural network to embed the spatial distribution of nuclei into their representations. To capture the correlations between the categories of nuclei and their surrounding tissue patterns, we further introduce edge features that are defined as the background textures between adjacent nuclei. Lastly, we integrate both polygon and graph structure learning mechanisms into a whole framework that can extract intra and inter-nucleus structural characteristics for nuclei classification. Experimental results show that the proposed framework achieves significant improvements compared to the previous methods. Code and data are made available via https://github.com/lhaof/SENC.

Characterization of sensory and motor dysfunction and morphological alterations in late stages of type 2 diabetic mice.

Diabetic neuropathy is the most common complication of diabetes and lacks effective treatments. Although sensory dysfunction during the early stages of diabetes has been extensively studied in various animal models, the functional and morphological alterations in sensory and motor systems during late stages of diabetes remain largely unexplored. In the current work, we examined the influence of diabetes on sensory and motor function as well as morphological changes in late stages of diabetes. The obese diabetic Leprdb/db mice (db/db) were used for behavioral assessments and subsequent morphological examinations. The db/db mice exhibited severe sensory and motor behavioral defects at the age of 32 weeks, including significantly higher mechanical withdrawal threshold and thermal latency of hindpaws compared with age-matched nondiabetic control animals. The impaired response to noxious stimuli was mainly associated with the remarkable loss of epidermal sensory fibers, particularly CGRP-positive nociceptive fibers. Unexpectedly, the area of CGRP-positive terminals in the spinal dorsal horn was dramatically increased in diabetic mice, which was presumably associated with microglial activation. In addition, the db/db mice showed significantly more foot slips and took longer time during the beam-walking examination compared with controls. Meanwhile, the running duration in the rotarod test was markedly reduced in db/db mice. The observed sensorimotor deficits and motor dysfunction were largely attributed to abnormal sensory feedback and muscle atrophy as well as attenuated neuromuscular transmission in aged diabetic mice. Morphological analysis of neuromuscular junctions (NMJs) demonstrated partial denervation of NMJs and obvious fragmentation of acetylcholine receptors (AChRs). Intrafusal muscle atrophy and abnormal muscle spindle innervation were also detected in db/db mice. Additionally, the number of VGLUT1-positive excitatory boutons on motor neurons was profoundly increased in aged diabetic mice as compared to controls. Nevertheless, inhibitory synaptic inputs onto motor neurons were similar between the two groups. This excitation-inhibition imbalance in synaptic transmission might be implicated in the disturbed locomotion. Collectively, these results suggest that severe sensory and motor deficits are present in late stages of diabetes. This study contributes to our understanding of mechanisms underlying neurological dysfunction during diabetes progression and helps to identify novel therapeutic interventions for patients with diabetic neuropathy.

Analysis of Fluid Replacement in Two Fluidic Chambers for Oblique-Incidence Reflectivity Difference (OI-RD) Biosensor.

Optical biosensors have a significant impact on various aspects of our lives. In many applications of optical biosensors, fluidic chambers play a crucial role in facilitating controlled fluid delivery. It is essential to achieve complete liquid replacement in order to obtain accurate and reliable results. However, the configurations of fluidic chambers vary across different optical biosensors, resulting in diverse fluidic volumes and flow rates, and there are no standardized guidelines for liquid replacement. In this paper, we utilize COMSOL Multiphysics, a finite element analysis software, to investigate the optimal fluid volume required for two types of fluidic chambers in the context of the oblique-incidence reflectivity difference (OI-RD) biosensor. We found that the depth of the fluidic chamber is the most crucial factor influencing the required liquid volume, with the volume being a quadratic function of the depth. Additionally, the required fluid volume is also influenced by the positions on the substrate surface bearing samples, while the flow rate has no impact on the fluid volume.

The multimodality cell segmentation challenge: toward universal solutions.

Cell segmentation is a critical step for quantitative single-cell analysis in microscopy images. Existing cell segmentation methods are often tailored to specific modalities or require manual interventions to specify hyper-parameters in different experimental settings. Here, we present a multimodality cell segmentation benchmark, comprising more than 1,500 labeled images derived from more than 50 diverse biological experiments. The top participants developed a Transformer-based deep-learning algorithm that not only exceeds existing methods but can also be applied to diverse microscopy images across imaging platforms and tissue types without manual parameter adjustments. This benchmark and the improved algorithm offer promising avenues for more accurate and versatile cell analysis in microscopy imaging.

SENSE: Self-Evolving Learning for Self-Supervised Monocular Depth Estimation.

Self-supervised depth estimation methods can achieve competitive performance using only unlabeled monocular videos, but they suffer from the uncertainty of jointly learning depth and pose without any ground truths of both tasks. Supervised framework provides robust and superior performance but is limited by the scope of the labeled data. In this paper, we introduce SENSE, a novel learning paradigm for self-supervised monocular depth estimation that progressively evolves the prediction result using supervised learning, but without requiring labeled data. The key contribution of our approach stems from the novel use of the pseudo labels - the noisy depth estimation from the self-supervised methods. We surprisingly find that a fully supervised depth estimation network trained using the pseudo labels can produce even better results than its "ground truth". To push the envelope further, we then evolve the self-supervised backbone by replacing its depth estimation branch with that fully supervised network. Based on this idea, we devise a comprehensive training pipeline that alternatively enhances the two key branches (depth and pose estimation) of the self-supervised backbone network. Our proposed approach can effectively ease the difficulty of multi-task training in self-supervised depth estimation. Experimental results have shown that our proposed approach achieves state-of-the-art results on the KITTI dataset.

Unbiased curriculum learning enhanced global-local graph neural network for protein thermodynamic stability prediction.

MOTIVATION: Proteins play crucial roles in biological processes, with their functions being closely tied to thermodynamic stability. However, measuring stability changes upon point mutations of amino acid residues using physical methods can be time-consuming. In recent years, several computational methods for protein thermodynamic stability prediction (PTSP) based on deep learning have emerged. Nevertheless, these approaches either overlook the natural topology of protein structures or neglect the inherent noisy samples resulting from theoretical calculation or experimental errors. RESULTS: We propose a novel Global-Local Graph Neural Network powered by Unbiased Curriculum Learning for the PTSP task. Our method first builds a Siamese graph neural network to extract protein features before and after mutation. Since the graph's topological changes stem from local node mutations, we design a local feature transformation module to make the model focus on the mutated site. To address model bias caused by noisy samples, which represent unavoidable errors from physical experiments, we introduce an unbiased curriculum learning method. This approach effectively identifies and re-weights noisy samples during the training process. Extensive experiments demonstrate that our proposed method outperforms advanced protein stability prediction methods, and surpasses state-of-the-art learning methods for regression prediction tasks. AVAILABILITY AND IMPLEMENTATION: All code and data is available at https://github.com/haifangong/UCL-GLGNN.

Detection speed optimization of the OI-RD microscope for ultra-high throughput screening.

The oblique-incidence reflectivity difference (OI-RD) microscope is a label-free detection system for microarrays that has many successful applications in high throughput drug screening. The increase and optimization of the detection speed of the OI-RD microscope will enable it to be a potential ultra-high throughput screening tool. This work presents a series of optimization methods that can significantly reduce the time to scan an OI-RD image. The wait time for the lock-in amplifier was decreased by the proper selection of the time constant and development of a new electronic amplifier. In addition, the time for the software to acquire data and for translation stage movement was also minimized. As a result, the detection speed of the OI-RD microscope is 10 times faster than before, making the OI-RD microscope suitable for ultra-high throughput screening applications.

Spatial correlated incidence modeling with zero inflation.

Disease mapping models have been popularly used to model disease incidence with spatial correlation. In disease mapping models, zero inflation is an important issue, which often occurs in disease incidence datasets with high proportions of zero disease count. It is originated from limited survey coverage or unadvanced testing equipment, which makes some regions have no observed patients. Then excessive zeros recorded in the disease incidence dataset would mess up the true distributions of disease incidence and lead to inaccurate estimates. To address this issue, a zero-inflated disease mapping model is developed in this work. In this model, a zero-inflated process using Bernoulli indicators is assumed to characterize whether the zero inflation occurs for each region. For regions without zero inflation, a coherent and generative disease mapping model is applied for mapping the spatially correlated disease incidence. Independent spatial random effects are incorporated in both processes to account for the spatial patterns of zero inflation and disease incidence. External covariates are also considered in both processes to better explain the disease count data. To estimate the model, a Markov chain Monte Carlo algorithm is proposed. We evaluate model performance via a variety of simulation experiments. Finally, a Lyme disease dataset of Virginia is analyzed to illustrate the application of the proposed model.

Thyroid region prior guided attention for ultrasound segmentation of thyroid nodules.

Ultrasound segmentation of thyroid nodules is a challenging task, which plays an vital role in the diagnosis of thyroid cancer. However, the following two factors limit the development of automatic thyroid nodule segmentation algorithms: (1) existing automatic nodule segmentation algorithms that directly apply semantic segmentation techniques can easily mistake non-thyroid areas as nodules, because of the lack of the thyroid gland region perception, the large number of similar areas in the ultrasonic images, and the inherently low contrast images; (2) the currently available dataset (i.e., DDTI) is small and collected from a single center, which violates the fact that thyroid ultrasound images are acquired from various devices in real-world situations. To overcome the lack of thyroid gland region prior knowledge, we design a thyroid region prior guided feature enhancement network (TRFE+) for accurate thyroid nodule segmentation. Specifically, (1) a novel multi-task learning framework that simultaneously learns the nodule size, gland position, and the nodule position is designed; (2) an adaptive gland region feature enhancement module is proposed to make full use of the thyroid gland prior knowledge; (3) a normalization approach with respect to the channel dimension is applied to alleviate the domain gap during the training process. To facilitate the development of thyroid nodule segmentation, we have contributed TN3K: an open-access dataset containing 3493 thyroid nodule images with high-quality nodule masks labeling from various devices and views. We perform a thorough evaluation based on the TN3K test set and DDTI to demonstrate the effectiveness of the proposed method. Code and data are available at https://github.com/haifangong/TRFE-Net-for-thyroid-nodule-segmentation.

Which Pixel to Annotate: A Label-Efficient Nuclei Segmentation Framework.

Recently deep neural networks, which require a large amount of annotated samples, have been widely applied in nuclei instance segmentation of H&E stained pathology images. However, it is inefficient and unnecessary to label all pixels for a dataset of nuclei images which usually contain similar and redundant patterns. Although unsupervised and semi-supervised learning methods have been studied for nuclei segmentation, very few works have delved into the selective labeling of samples to reduce the workload of annotation. Thus, in this paper, we propose a novel full nuclei segmentation framework that chooses only a few image patches to be annotated, augments the training set from the selected samples, and achieves nuclei segmentation in a semi-supervised manner. In the proposed framework, we first develop a novel consistency-based patch selection method to determine which image patches are the most beneficial to the training. Then we introduce a conditional single-image GAN with a component-wise discriminator, to synthesize more training samples. Lastly, our proposed framework trains an existing segmentation model with the above augmented samples. The experimental results show that our proposed method could obtain the same-level performance as a fully-supervised baseline by annotating less than 5% pixels on some benchmarks.

SSMD: Semi-Supervised medical image detection with adaptive consistency and heterogeneous perturbation.

Semi-Supervised classification and segmentation methods have been widely investigated in medical image analysis. Both approaches can improve the performance of fully-supervised methods with additional unlabeled data. However, as a fundamental task, semi-supervised object detection has not gained enough attention in the field of medical image analysis. In this paper, we propose a novel Semi-Supervised Medical image Detector (SSMD). The motivation behind SSMD is to provide free yet effective supervision for unlabeled data, by regularizing the predictions at each position to be consistent. To achieve the above idea, we develop a novel adaptive consistency cost function to regularize different components in the predictions. Moreover, we introduce heterogeneous perturbation strategies that work in both feature space and image space, so that the proposed detector is promising to produce powerful image representations and robust predictions. Extensive experimental results show that the proposed SSMD achieves the state-of-the-art performance at a wide range of settings. We also demonstrate the strength of each proposed module with comprehensive ablation studies.

Absolute quantitative analysis of endogenous neurotransmitters and amino acids by liquid chromatography-tandem mass spectrometry combined with multidimensional adsorption and collision energy defect.

Considering that neurotransmitters (NTs) and amino acids (AAs) exert pivotal roles in various neurological diseases, global detection of these endogenous metabolites is of great significance for the treatment of nervous system diseases. Herein, a workflow that could cope with various challenges was proposed to establish an extendable all-in-one injection liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay for analyzing these small molecular metabolites with high coverage. To obtain a qualified blank biological matrix for the preparation of standard curves and quality control samples, different absorption solvents, including activated carbon (AC), calcite (Cal) and montmorillonite (Mnt) were systematically evaluated for efficient absorption of endogenous substances with minimum residue. We also firstly proposed a "Collision Energy Defect (CED)" strategy to solve the huge difference of mass signal strength caused by different properties and concentrations of 11 NTs and 17 AAs. The quantitative results were validated by LC-MS/MS. Sensitivity, accuracy, and recovery meeting generally accepted bioanalytic guidelines were observed in a concentration span of at least 100 to 500 times for each analyte. Then the temporal changes of intracerebral and peripheral NTs and AAs in ischemic stroke model and sham operated rats were successfully produced and compared using the described method. All these results suggested that the currently developed assay was powerful enough to simultaneously monitor a large panel of endogenous small molecule metabolites, which was expected to be widely used in the research of various diseases mediated by NTs and AAs.

Depthwise Nonlocal Module for Fast Salient Object Detection Using a Single Thread.

Recently, deep convolutional neural networks have achieved significant success in salient object detection. However, existing state-of-the-art methods require high-end GPUs to achieve real-time performance, which makes it hard to adapt to low cost or portable devices. Although generic network architectures have been proposed to speed up inference on mobile devices, they are tailored to the task of image classification or semantic segmentation, and struggle to capture intrachannel and interchannel correlations that are essential for contrast modeling in salient object detection. Motivated by the above observations, we design a new deep-learning algorithm for fast salient object detection. The proposed algorithm for the first time achieves competitive accuracy and high inference efficiency simultaneously with a single CPU thread. Specifically, we propose a novel depthwise nonlocal module (DNL), which implicitly models contrast via harvesting intrachannel and interchannel correlations in a self-attention manner. In addition, we introduce a depthwise nonlocal network architecture that incorporates both DNLs module and inverted residual blocks. The experimental results show that our proposed network attains very competitive accuracy on a wide range of salient object detection datasets while achieving state-of-the-art efficiency among all existing deep-learning-based algorithms.

Online Alternate Generator against Adversarial Attacks.

The field of computer vision has witnessed phenomenal progress in recent years partially due to the development of deep convolutional neural networks. However, deep learning models are notoriously sensitive to adversarial examples which are synthesized by adding quasi-perceptible noises on real images. Some existing defense methods require to re-train attacked target networks and augment the train set via known adversarial attacks, which is inefficient and might be unpromising with unknown attack types. To overcome the above issues, we propose a portable defense method, online alternate generator, which does not need to access or modify the parameters of the target networks. The proposed method works by online synthesizing another image from scratch for an input image, instead of removing or destroying adversarial noises. To avoid pretrained parameters exploited by attackers, we alternately update the generator and the synthesized image at the inference stage. Experimental results demonstrate that the proposed defensive scheme and method outperforms a series of state-of-the-art defending models against gray-box adversarial attacks.

Ginsenoside Rb1 exerts neuroprotective effects through regulation of Lactobacillus helveticus abundance and GABAA receptor expression.

BACKGROUND: Ginsenoside Rb1 (Rb1), one of the most abundant protopanaxadiol-type ginsenosides, exerts excellent neuroprotective effects even though it has low intracephalic exposure. PURPOSE: The present study aimed to elucidate the apparent contradiction between the pharmacokinetics and pharmacodynamics of Rb1 by studying the mechanisms underlying neuroprotective effects of Rb1 based on regulation of microflora. METHODS: A pseudo germ-free (PGF) rat model was established, and neuroprotective effects of Rb1 were compared between conventional and PGF rats. The relative abundances of common probiotics were quantified to reveal the authentic probiotics that dominate in the neuroprotection of Rb1. The expressions of the gamma-aminobutyric acid (GABA) receptors, including GABAA receptors (α2, ß2, and γ2) and GABAB receptors (1b and 2), in the normal, ischemia/reperfusion (I/R), and I/R+Rb1 rat hippocampus and striatum were assessed to reveal the neuroprotective mechanism of Rb1. RESULTS: The results showed that microbiota plays a key role in neuroprotection of Rb1. The relative abundance of Lactobacillus helveticus (Lac.H) increased 15.26 fold after pretreatment with Rb1. I/R surgery induced effects on infarct size, neurological deficit score, and proinflammatory cytokines (IL-1ß, IL-6, and TNF-α) were prevented by colonizing the rat gastrointestinal tract with Lac.H (1 × 109 CFU) by gavage 15 d before I/R surgery. Both Rb1 and Lac.H upregulated expression of GABA receptors in I/R rats. Coadministration of a GABAA receptor antagonist significantly attenuated neuroprotective effects of Rb1 and Lac.H. CONCLUSION: In sum, Rb1 exerts neuroprotective effects by regulating Lac.H and GABA receptors rather than through direct distribution to the target sites.

ROSA: Robust Salient Object Detection Against Adversarial Attacks.

Recently, salient object detection has witnessed remarkable improvement owing to the deep convolutional neural networks which can harvest powerful features for images. In particular, the state-of-the-art salient object detection methods enjoy high accuracy and efficiency from fully convolutional network (FCN)-based frameworks which are trained from end to end and predict pixel-wise labels. However, such framework suffers from adversarial attacks which confuse neural networks via adding quasi-imperceptible noises to input images without changing the ground truth annotated by human subjects. To our knowledge, this paper is the first one that mounts successful adversarial attacks on salient object detection models and verifies that adversarial samples are effective on a wide range of existing methods. Furthermore, this paper proposes a novel end-to-end trainable framework to enhance the robustness for arbitrary FCN-based salient object detection models against adversarial attacks. The proposed framework adopts a novel idea that first introduces some new generic noise to destroy adversarial perturbations, and then learns to predict saliency maps for input images with the introduced noise. Specifically, our proposed method consists of a segment-wise shielding component, which preserves boundaries and destroys delicate adversarial noise patterns and a context-aware restoration component, which refines saliency maps through global contrast modeling. The experimental results suggest that our proposed framework improves the performance significantly for state-of-the-art models on a series of datasets.

Systematically identifying the hepatoprotective ingredients of schisandra lignan extract from pharmacokinetic and pharmacodynamic perspectives.

BACKGROUND: Herbal medicines (HMs) have been proven to be productive sources of leads for the development of drugs. To date approximately 150 lignans have been identified from Schisandra sphenanthera. Hepatoprotective activity is a well-known characteristic of schisandra lignans, yet the authentic types of active lignans are still not well known. PURPOSE: The present study aimed to develop a reliable and efficient strategy for identifying the hepatoprotective ingredients of schisandra lignan extract (SLE). METHODS: SLEs were prepared by extracting Schisandra sphenanthera powder using 10%, 50% and 90% ethanol (w/w 1:10) combining 5-fold volume of ethyl acetate. The schisandra lignans in SLEs were qualitatively analyzed based on liquid chromatography hybrid ion trap time-of-flight mass spectrometry (LCMS-IT-TOF). Preparative liquid chromatography (PLC) was used to collect ingredient fractions. The hepatoprotective activity of schisandra lignans was systematically investigated on in vivo and in vitro models. RESULTS: The SLE extracted by 50% ethanol and 5-fold volume of ethyl acetate (50%SLE) had the highest lignan content and exhibited significantly stronger hepatoprotective activity than other SLEs (P <  0.01). The hepatoprotective effect of 50%SLE mainly attributed to the SLE segment which collected from 12 to 22 min by PLC. Schisantherin A (Sth A) was confirmed as the most promising hepatoprotective drug in Schisandra sphenanthera due to high content in crude materials, high exposure level in vivo and high efficiency on APAP-induced hepatotoxicity. CONCLUSION: The hepatoprotective ingredients of SLEs were systematically investigated based on the presently developed approach, and Sth A was identified as the optimum hepatoprotective candidate in Schisandra sphenanthera.

Is the combinational administration of doxorubicin and glutathione a reasonable proposal?

The combinational administration of antioxidants and chemotherapeutic agents during conventional cancer treatment is among one of the most controversial areas in oncology. Although the data on the combinational usage of doxorubicin (DOX) and glutathione (GSH) agents have been explored for over 20 years, the duration, administration route, and authentic rationality have not yet been fully understood yet. In the current study, we systematically investigated the pharmacokinetics (PK) and pharmacodynamics (PD) with both in vivo and in vitro models to elucidate the influence of GSH on the toxicity and efficacy of DOX. We first studied the cardioprotective and hepatoprotective effects of GSH in Balb/c mice, H9c2, and HL7702 cells. We showed that coadministration of exogenous GSH (5, 50, and 500 mg/kg per day, intragastric) significantly attenuated DOX-induced cardiotoxicity and hepatotoxicity by increasing intracellular GSH levels, whereas the elevated GSH concentrations did not affect the exposure of DOX in mouse heart and liver. From PK and PD perspectives, then the influences of GSH on the chemotherapeutic efficacy of DOX were investigated in xenografted nude mice and cancer cell models, including MCF-7, HepG2, and Caco-2 cells, which revealed that administration of exogenous GSH dose-dependently attenuated the anticancer efficacy of DOX in vivo and in vitro, although the elevated GSH levels neither influenced the concentration of DOX in tumors in vivo, nor the uptake of DOX in MCF-7 tumor cells in vitro. Based on the results we suggest that the combined administration of GSH and DOX should be contraindicated during chemotherapy unless DOX has caused serious hepatotoxicity and cardiotoxicity.

Context-Aware Semantic Inpainting.

In recent times, image inpainting has witnessed rapid progress due to the generative adversarial networks (GANs) that are able to synthesize realistic contents. However, most existing GAN-based methods for semantic inpainting apply an auto-encoder architecture with a fully connected layer, which cannot accurately maintain spatial information. In addition, the discriminator in existing GANs struggles to comprehend high-level semantics within the image context and yields semantically consistent content. Existing evaluation criteria are biased toward blurry results and cannot well characterize edge preservation and visual authenticity in the inpainting results. In this paper, we propose an improved GAN to overcome the aforementioned limitations. Our proposed GAN-based framework consists of a fully convolutional design for the generator which helps to better preserve spatial structures and a joint loss function with a revised perceptual loss to capture high-level semantics in the context. Furthermore, we also introduce two novel measures to better assess the quality of image inpainting results. The experimental results demonstrate that our method outperforms the state-of-the-art under a wide range of criteria.