Lysosomal dysfunction and overload of nucleosides in thymidine phosphorylase deficiency of MNGIE.

Inherited deficiency of thymidine phosphorylase (TP), encoded by TYMP, leads to a rare disease with multiple mitochondrial DNA (mtDNA) abnormalities, mitochondrial neurogastrointestinal encephalomyopathy (MNGIE). However, the impact of TP deficiency on lysosomes remains unclear, which are important for mitochondrial quality control and nucleic acid metabolism. Muscle biopsy tissue and skin fibroblasts from MNGIE patients, patients with m.3243 A > G mitochondrial encephalopathy, lactic acidosis and stroke-like episodes (MELAS) and healthy controls (HC) were collected to perform mitochondrial and lysosomal functional analyses. In addition to mtDNA abnormalities, compared to controls distinctively reduced expression of LAMP1 and increased mitochondrial content were detected in the muscle tissue of MNGIE patients. Skin fibroblasts from MNGIE patients showed decreased expression of LAMP2, lowered lysosomal acidity, reduced enzyme activity and impaired protein degradation ability. TYMP knockout or TP inhibition in cells can also induce the similar lysosomal dysfunction. Using lysosome immunoprecipitation (Lyso- IP), increased mitochondrial proteins, decreased vesicular proteins and V-ATPase enzymes, and accumulation of various nucleosides were detected in lysosomes with TP deficiency. Treatment of cells with high concentrations of dThd and dUrd also triggers lysosomal dysfunction and disruption of mitochondrial homeostasis. Therefore, the results provided evidence that TP deficiency leads to nucleoside accumulation in lysosomes and lysosomal dysfunction, revealing the widespread disruption of organelles underlying MNGIE.

Point-Based Learnable Query Generator for Human-Object Interaction Detection.

Transformer-based and interaction point-based methods have demonstrated promising performance and potential in human-object interaction detection. However, due to differences in structure and properties, direct integration of these two types of models is not feasible. Recent Transformer-based methods divide the decoder into two branches: an instance decoder for human-object pair detection and a classification decoder for interaction recognition. While the attention mechanism within the Transformer enhances the connection between localization and classification, this paper focuses on further improving HOI detection performance by increasing the intrinsic correlation between instance and action features. To address these challenges, this paper proposes a novel Transformer-based HOI Detection framework. In the proposed method, the decoder contains three parts: learnable query generator, instance decoder, and interaction classifier. The learnable query generator aims to build an effective query to guide the instance decoder and interaction classifier to learn more accurate instance and interaction features. These features are then applied to update the query generator for the next layer. Especially, inspired by the interaction point-based HOI and object detection methods, this paper introduces the prior bounding boxes, keypoints detection and spatial relation feature to build the novel learnable query generator. Finally, the proposed method is verified on HICO-DET and V-COCO datasets. The experimental results show that the proposed method has the better performance compared with the state-of-the-art methods.

A Contrastive Learning Network for Performance Metric and Assessment of Physical Rehabilitation Exercises.

Human activity analysis in the legal monitoring environment plays an important role in the physical rehabilitation field, as it helps patients with physical injuries improve their postoperative conditions and reduce their medical costs. Recently, several deep learning-based action quality assessment (AQA) frameworks have been proposed to evaluate physical rehabilitation exercises. However, most of them treat this problem as a simple regression task, which requires both the action instance and its score label as input. This approach is limited by the fact that the annotations in this field usually consist of healthy or unhealthy labels rather than quality scores provided by professional physicians. Additionally, most of these methods cannot provide informative feedback on a patient's motion defects, which weakens their practical application. To address these problems, we propose a multi-task contrastive learning framework to learn subtle and critical differences from skeleton sequences to deal with the performance metric and AQA problems of physical rehabilitation exercises. Specifically, we propose a performance metric network that takes triplets of training samples as input for score generation. For the AQA task, the same contrast learning strategy is used, but pairwise training samples are fed into the action quality assessment network for score prediction. Notably, we propose quantifying the deviation of the joint attention matrix between different skeleton sequences and introducing it into the loss function of our learning network. It is proven that considering both score prediction loss and joint attention deviation loss improves physical exercises AQA performance. Furthermore, it helps to obtain informative feedback for patients to improve their motion defects by visualizing the joint attention matrix's difference. The proposed method is verified on the UI-PRMD and KIMORE datasets. Experimental results show that the proposed method achieves state-of-the-art performance.

Distinctive metabolic remodeling in TYMP deficiency beyond mitochondrial dysfunction.

Mitochondrial neurogastrointestinal encephalomyopathy (MNGIE) is caused by mutations in the TYMP gene, which encodes thymidine phosphorylase (TP). As a cytosolic metabolic enzyme, TP defects affect biological processes that are thought to not be limited to the abnormal replication of mitochondrial DNA. This study aimed to elucidate the characteristic metabolic alterations and associated homeostatic regulation caused by TYMP deficiency. The pathogenicity of novel TYMP variants was evaluated in terms of clinical features, genetic analysis, and structural instability. We analyzed plasma samples from three patients with MNGIE; three patients with m.3243A > G mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS); and four healthy controls (HC) using both targeted and untargeted metabolomics techniques. Transcriptomics analysis and bioenergetic studies were performed on skin fibroblasts from participants in these three groups. A TYMP overexpression experiment was conducted to rescue the observed changes. Compared with controls, specific alterations in nucleosides, bile acids, and steroid metabolites were identified in the plasma of MNGIE patients. Comparable mitochondrial dysfunction was present in fibroblasts from patients with TYMP deficiency and in those from patients with the m.3243A > G mutation. Distinctively decreased sterol regulatory element binding protein (SREBP) regulated cholesterol metabolism and fatty acid (FA) biosynthesis as well as reduced FA degradation were revealed in fibroblasts with TYMP deficiency. The restoration of thymidine phosphorylase activity rescued the observed changes in MNGIE fibroblasts. Our findings indicated that more widespread metabolic disturbance may be caused by TYMP deficiency in addition to mitochondrial dysfunction, which expands our knowledge of the biochemical outcome of TYMP deficiency. KEY MESSAGES: Distinct metabolic profiles in patients with TYMP deficiency compared to those with m.3243A > G mutation. TYMP deficiency leads to a global disruption of nucleoside metabolism. Cholesterol and fatty acid metabolism are inhibited in individuals with MNGIE. TYMP is functionally related to SREBP-regulated pathways. Potential metabolite biomarkers that could be valuable clinical tools to improve the diagnosis of MNGIE.

Online Multi-Label Streaming Feature Selection Based on Label Group Correlation and Feature Interaction.

Multi-label streaming feature selection has received widespread attention in recent years because the dynamic acquisition of features is more in line with the needs of practical application scenarios. Most previous methods either assume that the labels are independent of each other, or, although label correlation is explored, the relationship between related labels and features is difficult to understand or specify. In real applications, both situations may occur where the labels are correlated and the features may belong specifically to some labels. Moreover, these methods treat features individually without considering the interaction between features. Based on this, we present a novel online streaming feature selection method based on label group correlation and feature interaction (OSLGC). In our design, we first divide labels into multiple groups with the help of graph theory. Then, we integrate label weight and mutual information to accurately quantify the relationships between features under different label groups. Subsequently, a novel feature selection framework using sliding windows is designed, including online feature relevance analysis and online feature interaction analysis. Experiments on ten datasets show that the proposed method outperforms some mature MFS algorithms in terms of predictive performance, statistical analysis, stability analysis, and ablation experiments.

The incomplete circle of Willis is associated with vulnerable intracranial plaque features and acute ischemic stroke.

BACKGROUND: The circle of Willis (CoW) plays a significant role in intracranial atherosclerosis (ICAS). This study investigated the relationship between different types of CoW, atherosclerosis plaque features, and acute ischemic stroke (AIS). METHODS: We investigated 97 participants with AIS or transient ischemic attacks (TIA) underwent pre- and post-contrast 3T vessel wall cardiovascular magnetic resonance within 7 days of the onset of symptoms. The culprit plaque characteristics (including enhancement grade, enhancement ratio, high signal in T1, irregularity of plaque surface, and normalized wall index), and vessel remodeling (including arterial remodeling ratio and positive remodeling) for lesions were evaluated. The anatomic structures of the anterior and the posterior sections of the CoW (A-CoW and P-CoW) were also evaluated. The plaque features were compared among them. The plaque features were also compared between AIS and TIA patients. Finally, univariate and multivariate regression analysis was performed to evaluate the independent risk factors for AIS. RESULT: Patients with incomplete A-CoW showed a higher plaque enhancement ratio (P = 0.002), enhancement grade (P = 0.01), and normalized wall index (NWI) (P = 0.018) compared with the patients with complete A-CoW. A higher proportion of patients with incomplete symptomatic P-CoW demonstrated more culprit plaques with high T1 signals (HT1S) compared with those with complete P-CoW (P = 0.013). Incomplete A-CoW was associated with a higher enhancement grade of the culprit plaques [odds ratio (OR):3.84; 95% CI: 1.36-10.88, P = 0.011], after adjusting for clinical risk factors such as age, sex, smoking, hypertension, hyperlipemia, and diabetes mellitus. Incomplete symptomatic P-CoW was associated with a higher probability of HT1S (OR:3.88; 95% CI: 1.12-13.47, P = 0.033), after adjusting for clinical risk factors such as age, sex, smoking, hypertension, hyperlipemia, and diabetes mellitus. Furthermore, an irregularity of the plaque surface (OR: 6.24; 95% CI: 2.25-17.37, P < 0.001), and incomplete symptomatic P-CoW (OR: 8.03, 95% CI: 2.43-26.55, P = 0.001) were independently associated with AIS. CONCLUSIONS: This study demonstrated that incomplete A-CoW was associated with enhancement grade of the culprit plaque, and incomplete symptomatic side P-CoW was associated with the presence of HT1S of culprit plaque. Furthermore, an irregularity of plaque surface and incomplete symptomatic side P-CoW were associated with AIS.

Label-reconstruction-based pseudo-subscore learning for action quality assessment in sporting events.

Most existing action quality assessment (AQA) methods provide only an overall quality score for the input video and lack an evaluation of each substage of the movement process; thus, these methods cannot provide detailed feedback for users. Moreover, the existing datasets do not provide labels for substage quality assessment. To address these problems, in this work, a new label-reconstruction-based pseudo-subscore learning (PSL) method is proposed for AQA in sporting events. In the proposed method, the overall score of an action is not only regarded as a quality label but also used as a feature of the training set. A label-reconstruction-based learning algorithm is built to generate pseudo-subscore labels for the training set. Moreover, based on the pseudo-subscore labels and overall score labels, a multi-substage AQA model is fine-tuned from the PSL model to predict the action quality score of each substage and the overall score for an athlete. Several ablation experiments are performed to verify the effectiveness of each module. The experimental results show that our approach achieves state-of-the-art performance.

An ionic lock and a hydrophobic zipper mediate the coupling between an insect pheromone receptor BmOR3 and downstream effectors.

Pheromone receptors (PRs) recognize specific pheromone compounds to guide the behavioral outputs of insects, which are the most diverse group of animals on earth. The activation of PRs is known to couple to the calcium permeability of their coreceptor (Orco) or putatively with G proteins; however, the underlying mechanisms of this process are not yet fully understood. Moreover, whether this transverse seven transmembrane domain (7TM)-containing receptor is able to couple to arrestin, a common effector for many conventional 7TM receptors, is unknown. Herein, using the PR BmOR3 from the silk moth Bombyx mori and its coreceptor BmOrco as a template, we revealed that an agonist-induced conformational change of BmOR3 was transmitted to BmOrco through transmembrane segment 7 from both receptors, resulting in the activation of BmOrco. Key interactions, including an ionic lock and a hydrophobic zipper, are essential in mediating the functional coupling between BmOR3 and BmOrco. BmOR3 also selectively coupled with Gi proteins, which was dispensable for BmOrco coupling. Moreover, we demonstrated that trans-7TM BmOR3 recruited arrestin in an agonist-dependent manner, which indicates an important role for BmOR3-BmOrco complex formation in ionotropic functions. Collectively, our study identified the coupling of G protein and arrestin to a prototype trans-7TM PR, BmOR3, and provided important mechanistic insights into the coupling of active PRs to their downstream effectors, including coreceptors, G proteins, and arrestin.

Novel biallelic mutations in POLG gene: large deletion and missense variant associated with PEO.

BACKGROUND: Mitochondrial disorders are clinically heterogeneous diseases associated with impaired oxidative phosphorylation (OXPHOS) activity. POLG, which encodes the DNA polymerase-γ (Polγ) catalytic subunit, is the most commonly mutated nuclear gene associated with mitochondrial disorders. METHODS: We carried out whole-exome sequencing (WES) to identify the gene associated with progressive external ophthalmoplegia (PEO). We then performed histopathological analyses, assessed mitochondrial biology, and executed functional studies to evaluate the potential pathogenicity of the identified genetic mutations. RESULTS: Novel biallelic POLG mutations, including a large deletion mutation (exons 7-21) and a missense variant c.1796C>T (p.Thr599Ile) were detected in the proband. Histopathological analysis of a biopsied muscle sample from this patient revealed the presence of approximately 20% COX-negative fibers. Bioinformatics analyses confirmed that the detected mutations were pathogenic. Furthermore, levels of mitochondrial complex I, II, and IV subunit protein expressions were found to be decreased in the proband, and marked impairment of mitochondrial respiration was evident in cells harboring these mutations. CONCLUSION: This study expands the spectrum of known POLG variants associated with PEO and advances current understanding regarding the structural and functional impacts of these mutations.

Functional suppression of Ripk1 blocks the NF-κB signaling pathway and induces neuron autophagy after traumatic brain injury.

Traumatic brain injury (TBI), known as intracranial injury, has been a serious threat to human health. Evidence exists indicating that autophagy and inflammatory responses contribute to secondary brain injury after TBI. Notably, receptor-interacting protein kinase 1 (Ripk1) exerts an important role in cell autophagy. Therefore, this study aims to explore the effect of Ripk1 on neuron autophagy and apoptosis in TBI. Initially, blood samples of patients with TBI and healthy persons were collected to detect the expression of Ripk1, nuclear factor-kappa B (NF-κB), and NF-kB inhibitor α (IKBα). Then rat models with TBI were successfully established and, respectively, treated with shRNA targeting Ripk1 (sh-Ripk1), Ripk1 overexpression plasmid (oe-Ripk1), or IKKα inhibitor (BAY 11-7082). Subsequently, reverse transcription quantitative polymerase chain reaction and Western blot analysis were conducted to detect the expression of Ripk1, IKBα, NF-κB signaling pathway-, and apoptosis-related factors. Enzyme-linked immunosorbent assay was used to detect the expression of inflammatory cytokines. Compared with healthy persons, the expression of Ripk1, NF-κB and IKBα in blood of TBI patients was significantly upregulated. After silencing of Ripk1 or inhibition of the NF-κB signaling pathway, the expression of IL-1ß, IL-6, TNF-α, Bax, and cleaved-caspase-3 was downregulated, and the expression of Bcl-2, ATG5, and LC3II/LC3I was upregulated. Furthermore, neuron injury and apoptosis were notably reduced and neuron autophagy increased significantly by Ripk1 downregulation or IKKα inhibitor. Ripk1 overexpression contributed to activation of NF-κB signaling pathway, whereby aggravating TBI-induced damage. Silencing Ripk1 suppresses TBI by inhibiting inflammation and promoting autophagy of neurons via inhibition of NF-κB signaling pathway.

A novel m.11406 T > A mutation in mitochondrial ND4 gene causes MELAS syndrome.

Pathogenic point mutations of mitochondrial DNA (mtDNA) are associated with a large number of heterogeneous diseases involving multiple systems with which patients may present with a wide range of clinical phenotypes. In this study, we describe a novel heteroplasmic missense mutation, m.11406 T > A, of the ND4 gene encoding the subunit 4 of mitochondrial complex I in a 32-year-old woman with recurrent epileptic seizure, headache and bilateral hearing loss. Skeletal muscle histochemistry demonstrated that approximately 20% of fibers were cytochrome C oxidase (COX) deficient with increased activity of succinate dehydrogenase (SDH). Further investigations in muscle specimens showed significantly reduced level of ND4 protein. It is interesting that the subunits of complex I (ND1 and NDFUB8) and complex IV(CO1) were also remarkably decreased. These findings indicate that ND1, NDFUB8 and CO1 are more susceptible than other subunits to mutations in the mitochondrial ND4 gene.

A Survey of Vision-Based Human Action Evaluation Methods.

The fields of human activity analysis have recently begun to diversify. Many researchers have taken much interest in developing action recognition or action prediction methods. The research on human action evaluation differs by aiming to design computation models and evaluation approaches for automatically assessing the quality of human actions. This line of study has become popular because of its explosively emerging real-world applications, such as physical rehabilitation, assistive living for elderly people, skill training on self-learning platforms, and sports activity scoring. This paper presents a comprehensive survey of approaches and techniques in action evaluation research, including motion detection and preprocessing using skeleton data, handcrafted feature representation methods, and deep learning-based feature representation methods. The benchmark datasets from this research field and some evaluation criteria employed to validate the algorithms' performance are introduced. Finally, the authors present several promising future directions for further studies.

A Comprehensive Survey of Vision-Based Human Action Recognition Methods.

Although widely used in many applications, accurate and efficient human action recognition remains a challenging area of research in the field of computer vision. Most recent surveys have focused on narrow problems such as human action recognition methods using depth data, 3D-skeleton data, still image data, spatiotemporal interest point-based methods, and human walking motion recognition. However, there has been no systematic survey of human action recognition. To this end, we present a thorough review of human action recognition methods and provide a comprehensive overview of recent approaches in human action recognition research, including progress in hand-designed action features in RGB and depth data, current deep learning-based action feature representation methods, advances in human⁻object interaction recognition methods, and the current prominent research topic of action detection methods. Finally, we present several analysis recommendations for researchers. This survey paper provides an essential reference for those interested in further research on human action recognition.

Intelligent diagnosis with Chinese electronic medical records based on convolutional neural networks.

BACKGROUND: Benefiting from big data, powerful computation and new algorithmic techniques, we have been witnessing the renaissance of deep learning, particularly the combination of natural language processing (NLP) and deep neural networks. The advent of electronic medical records (EMRs) has not only changed the format of medical records but also helped users to obtain information faster. However, there are many challenges regarding researching directly using Chinese EMRs, such as low quality, huge quantity, imbalance, semi-structure and non-structure, particularly the high density of the Chinese language compared with English. Therefore, effective word segmentation, word representation and model architecture are the core technologies in the literature on Chinese EMRs. RESULTS: In this paper, we propose a deep learning framework to study intelligent diagnosis using Chinese EMR data, which incorporates a convolutional neural network (CNN) into an EMR classification application. The novelty of this paper is reflected in the following: (1) We construct a pediatric medical dictionary based on Chinese EMRs. (2) Word2vec adopted in word embedding is used to achieve the semantic description of the content of Chinese EMRs. (3) A fine-tuning CNN model is constructed to feed the pediatric diagnosis with Chinese EMR data. Our results on real-world pediatric Chinese EMRs demonstrate that the average accuracy and F1-score of the CNN models are up to 81%, which indicates the effectiveness of the CNN model for the classification of EMRs. Particularly, a fine-tuning one-layer CNN performs best among all CNNs, recurrent neural network (RNN) (long short-term memory, gated recurrent unit) and CNN-RNN models, and the average accuracy and F1-score are both up to 83%. CONCLUSION: The CNN framework that includes word segmentation, word embedding and model training can serve as an intelligent auxiliary diagnosis tool for pediatricians. Particularly, a fine-tuning one-layer CNN performs well, which indicates that word order does not appear to have a useful effect on our Chinese EMRs.

Axially Symmetric Data Clustering Through Dirichlet Process Mixture Models of Watson Distributions.

This paper proposes a Bayesian nonparametric framework for clustering axially symmetric data. Our approach is based on a Dirichlet processes mixture model with Watson distributions, which can also be considered as the infinite Watson mixture model. In this paper, first, we extend the finite Watson mixture model into its infinite counterpart based on the framework of truncated Dirichlet process mixture model with a stick-breaking representation. Second, we propose a coordinate ascent mean-field variational inference algorithm that can effectively learn the parameters of our model with closed-form solutions; Third, to cope with a massive data set, we develop a stochastic variational inference algorithm to learn the proposed model through the method of stochastic gradient ascent; Finally, the proposed nonparametric Bayesian model is evaluated through simulated axially symmetric data sets and a real-world application, namely, gene expression data clustering.

MicroRNA-139-5p/Flt1/Wnt/ß-catenin regulatory crosstalk modulates the progression of glioma.

Fms-related tyrosine kinase 1 (Flt1), the receptor of VEGF/PIGF, is associated with cancer angiogenesis and tumorigenesis. Although the high expression of Flt1 in glioma is identified, its regulatory mechanism remains unclear. In the present study, we demonstrate that miR­139­5p inhibits Flt1 expression mediated by binding its 3' untranslated region (3'UTR) to regulate the progression of human glioma. We found miR­139­5p was downregulated in glioma tissues compared with normal brain tissues whereas a converse expression level of Flt1 was observed. Additionally we proved that miR­139­5p directly integrated with the 3'UTR of Flt1 via luciferase activity assay and cells transfected with miR­139­5p characterized with a low expression of Flt1 in mRNA and protein levels. Furthermore, we validated that miR­139­5p enforced its biological modulation via targeting Flt1 through rescue experiments. miR­139­5p suppressed and Flt1 stimulated the malignant activities of glioma cells. We demonstrated that miR­139­5p inhibited the Flt1-mediated Wnt/ß-catenin signaling pathway in glioma cells. Conclusively, our study indicated that miR­139­5p can counteract the malignant phenotypes of glioma cells by the inhibitory effect of the Flt1-mediated Wnt/ß-catenin signaling pathway.

MiR-433-3p suppresses cell growth and enhances chemosensitivity by targeting CREB in human glioma.

Previous studies reported that miR-433 exerts function widely in human tumorigenesis and development. Here, we further investigate the potential role of miR-433 in glioma. Quantitative real-time PCR demonstrated that miR-433-3p and miR-433-5p were low expressed in glioma tissues and cell lines. Functional studies suggested that the overexpression of miR-433-3p suppressed proliferation, induced apoptosis and inhibited invasion and migration of human glioma cells. But the growth and metastasis of glioma cells were not significantly influenced by overexpression of miR-433-5p. In a xenograft model, we also showed that miR-433-3p had an inhibitory effect on the growth of glioma. Bioinformatics coupled with luciferase and western blot assays revealed that CREB is a direct target of miR-433-3p, and the overexpression of CREB can rescue the phenotype changes induced by miR-433-3p overexpression. Besides, miR-433-3p could increase chemosensitivity of glioma to temozolomide by targeting CREB in vitro and in vivo. Taken together, these results suggest that miR-433-3p may function as a potential marker in diagnostic and therapeutic target for glioma.

Convolutional Deep Belief Networks for Single-Cell/Object Tracking in Computational Biology and Computer Vision.

In this paper, we propose deep architecture to dynamically learn the most discriminative features from data for both single-cell and object tracking in computational biology and computer vision. Firstly, the discriminative features are automatically learned via a convolutional deep belief network (CDBN). Secondly, we design a simple yet effective method to transfer features learned from CDBNs on the source tasks for generic purpose to the object tracking tasks using only limited amount of training data. Finally, to alleviate the tracker drifting problem caused by model updating, we jointly consider three different types of positive samples. Extensive experiments validate the robustness and effectiveness of the proposed method.

Robust Individual-Cell/Object Tracking via PCANet Deep Network in Biomedicine and Computer Vision.

Tracking individual-cell/object over time is important in understanding drug treatment effects on cancer cells and video surveillance. A fundamental problem of individual-cell/object tracking is to simultaneously address the cell/object appearance variations caused by intrinsic and extrinsic factors. In this paper, inspired by the architecture of deep learning, we propose a robust feature learning method for constructing discriminative appearance models without large-scale pretraining. Specifically, in the initial frames, an unsupervised method is firstly used to learn the abstract feature of a target by exploiting both classic principal component analysis (PCA) algorithms with recent deep learning representation architectures. We use learned PCA eigenvectors as filters and develop a novel algorithm to represent a target by composing of a PCA-based filter bank layer, a nonlinear layer, and a patch-based pooling layer, respectively. Then, based on the feature representation, a neural network with one hidden layer is trained in a supervised mode to construct a discriminative appearance model. Finally, to alleviate the tracker drifting problem, a sample update scheme is carefully designed to keep track of the most representative and diverse samples during tracking. We test the proposed tracking method on two standard individual cell/object tracking benchmarks to show our tracker's state-of-the-art performance.

Multi-surface analysis for human action recognition in video.

The majority of methods for recognizing human actions are based on single-view video or multi-camera data. In this paper, we propose a novel multi-surface video analysis strategy. The video can be expressed as three-surface motion feature (3SMF) and spatio-temporal interest feature. 3SMF is extracted from the motion history image in three different video surfaces: horizontal-vertical, horizontal- and vertical-time surface. In contrast to several previous studies, the prior probability is estimated by 3SMF rather than using a uniform distribution. Finally, we model the relationship score between each video and action as a probability inference to bridge the feature descriptors and action categories. We demonstrate our methods by comparing them to several state-of-the-arts action recognition benchmarks.