Tao Li

Nanjing University of Posts and Telecommunications, Nan-ching, Jiangsu, China

Are you Tao Li?

Claim your profile

Publications (424)428.9 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Age-related macular degeneration (AMD) is the leading cause of blindness among senior citizens of developed countries, with currently unknown etiology. Despite the close associations between AMD development and inhibitory complement factor H mutations, the first step of complement activation, which is the antibody response in AMD patients, has not been studied. Here, we obtained blood and tear samples from AMD patients and Non-AMD controls. We found that compared to Non-AMD controls, AMD subjects had increased IgA titers in serum and tear, and had elevated levels of circulating antibody-secreting plasmablasts. The increase in antibody titer was limited to the IgA isotype, since no significant differences were observed in IgM and IgG isotypes between AMD patients and Non-AMD controls. Interestingly, this increased antibody response in AMD patients was correlated with disease severity, as late AMD patients had increased IgA titers in serum and tear, as well as elevated plasmablast frequency after staphylococcal enterotoxin B stimulation, compared to early AMD patients. Together, our results implicated a role of overreactive IgA responses in AMD pathogenesis.
    No preview · Article · Jan 2016 · Immunobiology
  • Source
    Kun Cai · Wei Tu · Yuenan Liu · Tao Li · Hui Wang
    [Show abstract] [Hide abstract]
    ABSTRACT: Infection with Escherichia coli O157:H7 may develop into hemorrhagic colitis, or hemolytic uremic syndrome (HUS), which usually causes kidney failure or even death. The adhesion and toxins are the important virulent factors. In this study, a novel vaccine candidate rSOBGs was constructed based on the bacterial ghost (BG). rSOBGs maintained the integrity of cellular morphology and displayed the linear Stx2Am-Stx1B antigen on the surface of outer membrane. rSOBGs induced Stxs-specific IgA/IgG antibodies and stronger intimin-specific IgA/IgG antibodies effectively in sera in this study. In vivo, the rSOBGs provided the higher protection rate (52%) than native bacterial ghost-OBGs (12%) when challenged intragastricly with high dose (500 LD50) viable E. coli O157:H7. Meanwhile, the rSOBGs provided higher protection rate (73.33%) than OBGs when challenged with 2 LD50 even to 5 LD50 lysed E. coli O157:H7. In vitro, the rSOBGs-immunized sera possessed neutralizing activity to lysed pathogenic bacteria. Furthermore, the results of histopathology also displayed that the administration of rSOBGs have the ability to reduce or inhibit the adhesion lesions and toxins damages of organs. The novel vaccine candidate rSOBGs induced both anti-toxin and anti-adhesion immune protection, suggesting the possibility to prevent the infectious diseases caused by Escherichia coli O157:H7.
    Preview · Article · Dec 2015 · Scientific Reports
  • Jian Xu · Liang Tang · Chunqiu Zeng · Tao Li
    [Show abstract] [Hide abstract]
    ABSTRACT: Pattern discovery is one of the most fundamental problems in data mining. Various patterns with their discovering algorithms are proposed in different applications and domains. There is still a great demand for defining new meaningful patterns with new requirements since every application has its unique characteristics. Existing studies propose new query languages to describe these ad-hoc patterns. However, most of them focus on small variations of frequent item sets and association rules. Many meaningful patterns in other domains, such as temporal and spatial patterns, are not covered. This paper proposes a constraint based view for pattern discovery without introducing new languages, where the patterns are described by a collection of constraints given at run time. In this view, a pattern discovery problem is seen as a constraint satisfaction problem. This view provides a general framework for universal pattern discovery. Many previously known patterns can be regarded as different variations derived from this general framework with different constraints. Two generic algorithms are proposed for solving the constraint satisfaction problem. Empirical evaluation on two well-studied patterns shows that (1) the time cost of one generic algorithm is close to that of those specialized mining algorithms, and (2) the space cost of the generic algorithm increases linearly according to the input data volume. Two other case studies also demonstrate the effectiveness of this constraint based view for solving new problems in new scenarios.
    No preview · Article · Dec 2015
  • Qifeng Zhou · Hao Zhou · Tao Li
    [Show abstract] [Hide abstract]
    ABSTRACT: Feature selection aims to select a small subset of informative features that contain most of the information related to a given task. Existing feature selection methods often assume that all the features have the same cost. However, in many real world applications, different features may have different costs (e.g., different tests a patient might take in medical diagnosis). Ignoring the feature cost may produce good feature subsets in theory but they can not be used in practice. In this paper, we propose a random forest-based feature selection algorithm that incorporates the feature cost into the base decision tree construction process to produce low-cost feature subsets. In particular, when constructing a base tree, a feature is randomly selected with a probability inversely proportional to its associated cost. We evaluate the proposed method on a number of UCI datasets and apply it to a medical diagnosis problem where the real feature costs are estimated by experts. The experimental results demonstrate that our feature-cost-sensitive random forest (FCS-RF) is able to select a low-cost subset of informative features and achieves better performance than other state-of-art feature selection methods in real-world problems.
    No preview · Article · Dec 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To document a new clinical manifestation in familial exudative vitreoretinopathy (FEVR) eyes, especially in mild asymptomatic eyes with normal vision. Methods: Twenty individuals with mild Stage I or II FEVR with a conventional "normal-appearing" posterior pole and 20 healthy control eyes were recruited. The crossing numbers of retinal vessels with peripapillary inner reference circle, peripapillary outer reference circle, peripapillary temporal inner arc, peripapillary temporal outer arc, and branching points between the peripapillary outer reference circle and peripapillary inner reference circle were counted. Vessel bifurcation was evaluated by B/CI (defined as the branching number divided by the crossing number on peripapillary inner reference circle) and CO/B (crossing number on peripapillary outer reference circle divided by the branching number) ratios. The inter- and intraobservers' agreements were analyzed. All these parameters were compared between FEVR and control groups. Results: The coefficient of repeatability for the parameters ranged from 2.597 to 5.096, and the intraclass correlation coefficients were all above 0.85. All the parameters showed good interobserver agreement with a narrow range of 95% limit of agreement (from -3.16 to 3.37) and high Pearson correlation (P < 0.001). The mean crossing numbers on peripapillary inner reference circle, peripapillary outer reference circle, peripapillary temporal inner arc, peripapillary temporal outer arc, and the branching numbers were larger in the FEVR group. No significant differences were found in CO/B and B/CI ratios. Conclusion: Patients with FEVR have more vessels radiated from the optic disk in the posterior pole. Unlike the increased vessels in the peripheral retina, the increasing pattern of peripapillary vascularity in patients with FEVR does not appear to have a component of overbifurcation. This is a new documented clinical manifestation in patients with FEVR. Attention to an increased or arrangement pattern of retinal vessels may aid in the screening of FEVR.
    No preview · Article · Dec 2015 · Retina
  • [Show abstract] [Hide abstract]
    ABSTRACT: Social media has become increasingly prevalent in the last few years, not only enabling people to connect with each other by social links, but also providing platforms for people to share information and interact over diverse topics. Rich user-generated information, for example, users relationships and daily posts, are often available in most social media service websites. Given such information, a challenging problem is to provide reasonable user and community recommendation for a target user, and consequently, help the target user engage in the daily discussions and activities with his/her friends or like-minded people. In this article, we propose a unified framework of recommending users and communities that utilizes the information in social media. Given a users profile or a set of keywords as input, our framework is capable of recommending influential users and topic-cohesive interactive communities that are most relevant to the given user or keywords. With the proposed framework, users can find other individuals or communities sharing similar interests, and then have more interaction with these users or within the communities. We present a generative topic model to discover user-oriented and community-oriented topics simultaneously, which enables us to capture the exact topical interests of users, as well as the focuses of communities. Extensive experimental evaluation and case studies on a dataset collected from Twitter demonstrate the effectiveness of our proposed framework compared with other probabilistic-topic-model-based recommendation methods.
    No preview · Article · Oct 2015 · ACM Transactions on Knowledge Discovery from Data
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective: To analyze the risk factors of neurological complications of posterior vertebral column resection in the treatment of severe rigid congenital spinal deformities. Methods: The clinical data of 88 patients with severe rigid congenital spinal deformities who underwent PVCR in Department Of Orthopaedics, Xijing Hospital, Fourth Military Medical University from June 2007 to November 2012 were collected. There were 39 males and 49 females at the average age of 16.9 years (range 6-46 years). To measure the Cobb angle and balance at preoperative, postoperative and follow up, and to record the operation report, neurological complications and at follow up. The relevant factors of neurological complications were analyzed by one-way analysis, including: age, Cobb angle, operation time, body mass index, pulmonary function, blood volume loss, resection level, number of vertebrae fixed, number of vertebrae resected, usage of cage or titanium mesh, preoperative neurologic function, the type of deformity and combination of spinal canal deformity, and further analyzed by multiariable Logistic regression analysis. Results: The average follow up was 42 months (range 19 to 83 months). The number of resected vertebrae average 1.3 (range 1 to 3), operative time average 502.4 min (range 165.0 to 880.0 min), estimate blood loss average 2 238 ml (range 100 to 11 500 ml) for an average 69.3% blood volume loss (range 9% to 299%). The average preoperative major coronal curve of 93.6° corrected to 22.2°, at the final follow-up, the coronal curve was 22.2° with a correction of 76.8%. The average preoperative coronal imbalance (absolute value) was 2.5 cm decreasing to 1.3 cm at the final follow-up. The average preoperative major sagittal curve of 88.2° corrected to 28.7°, at the final follow-up, the sagittal curve was 29.2°, average decrease in kyphosis of 59.0°. The average preoperative sagittal imbalance (absolute value) was 3.1 cm decreasing to 1.2 cm at the final follow-up. There were 12 patients (13.6%) developed a neurological complications. High rate of neurological complications was occurred in patients with operative time greater than 480 min, pulmonary dysfunction, blood volume loss greater than 50%, T7 - T99 osteotomy and preoperative neurologic compromise (P=0.046, 0.000, 0.000, 0.033, 0.043). Conclusions: Posterior vertebral column resection can achieve satisfactory efficacy in treatment of severe spinal deformities. Pulmonary dysfunction and blood volume loss greater than 50% were significant risk factors of neurological complications.
    No preview · Article · Sep 2015 · Zhonghua wai ke za zhi [Chinese journal of surgery]
  • Article: Soter
    Yanfang Ye · Tao Li · Haiyin Shen

    No preview · Article · Jul 2015 · ACM Transactions on Intelligent Systems and Technology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The present study aimed to investigate the ability of SS31, a novel mitochondria‑targeted peptide to protect against t‑BHP‑induced mitochondrial dysfunction and apoptosis in 661W cell lines. The 661W cells were treated with various concentrations of SS‑31 and an MTT assay was used to determine cell viability. The expression of nitrotyrosine and 8‑hydroxydeoxyguanosine (8‑OHdG) was detected using immunofluorescent staining. Apoptosis were assessed using Hoechst staining and an annexin V/propidium iodide flow cytometer. Reactive oxygen species (ROS) were detected using MitoSOXTM with confocal microscopy. Changes in mitochondrial membrane potential were analyzed using flow cytometry. In addition, the release of cytochrome c was analyzed using confocal microscopy. The viability of the cells improved following treatment with SS31 between 100 nM and 1 µM, compared with untreated control group. Compared with the t‑BHP treatment group (20.0±3.8%), the number of annexin V‑positive cells decreased dose‑dependently to 13.6±2.6, 9.8±0.5 and 7.4±2.0% in the SS‑31 treated group at concentrations of 10 nM, 100 nM and 1 µM, respectively. Treatment with SS‑31 significantly prevented the t‑BHP‑induced expression of nitrotyrosine and 8‑OHdG, decreased the quantity of mitochondrial ROS, increased mitochondrial potential, and prevented the release of cytochrome c from mitochondria into the cytoplasm. Therefore, the SS31 mitochondria‑targeted peptide protected the 661W cells from the sustained oxidative stress induced by t‑BHP.
    Preview · Article · Jul 2015 · Molecular Medicine Reports
  • [Show abstract] [Hide abstract]
    ABSTRACT: An intelligent detection method is proposed in this paper to enrich the study of applying machine learning and data mining techniques to building structural damage identification. The proposed method integrates the multi-sensory data fusion and classifier ensemble to detect the location and extent of the damage. First, the wavelet package analysis is used to transform the original vibration acceleration signal into energy features. Then the posteriori probability support vector machines (PPSVM) and the Dempster-Shafer (DS) evidence theory are combined to identify the damage. Empirical study on a benchmark structure model shows that, compared with popular data mining approaches, the proposed method can provide more accurate and stable detection results. Furthermore, this paper compares the detection performance of the information fusion at different levels. The experimental analysis demonstrates that the proposed method with the fusion at the decision level can make good use of multi-sensory information and is more robust in practice.
    No preview · Article · Jul 2015 · Applied Soft Computing

  • No preview · Article · Jun 2015 · The spine journal: official journal of the North American Spine Society
  • Linnan Wang · Tao Li · Yueming Song

    No preview · Article · Jun 2015 · The spine journal: official journal of the North American Spine Society
  • [Show abstract] [Hide abstract]
    ABSTRACT: In many online classification tasks or non-exhaustive learning, it is often impossible to define a training set with a complete set of classes. The presence of new classes as well as the novelties caused by data errors can severely affect the performance of classifiers. Traditional proximity-based approaches usually utilize the distance to measure the proximity of different samples. In this study, we propose a framework that uses ensemble learning to detect novelty based on Random Forest (RF). The proposed framework is based on the observation that an ensemble of classifiers can provide a kind of metric to characterize different classes and measure their proximity. In particular, we apply ensemble methods with the decision tree as base classifiers and present two specific approaches, RFV and RFP, based on random forest. RFV uses the vote distribution of RF on a testing sample, and RFP takes the proximity matrix of RF as a special kernel metric to discover the novelty. The proposed approaches are compared against two common approaches: support vector domain description (SVDD) and Gaussian Mixed Model (GMM) on one artificial data set and five benchmark data sets. The experimental results show that the proposed methods achieve better performance in terms of accuracy and recall.
    No preview · Article · Jun 2015 · Expert Systems with Applications
  • Longhui Zhang · Lei Li · Tao Li

    No preview · Article · May 2015 · ACM SIGKDD Explorations Newsletter
  • Source
    Jian Xu · Yexi Jiang · Chunqiu Zeng · Tao Li
    [Show abstract] [Hide abstract]
    ABSTRACT: Identifying the anomalies is a critical task to maintain the uptime of the monitored distributed systems. For this reason, the trace data collected from real time monitors are often provided in form of streams for anomaly detection. Due to the dramatic increase of the scale of modern distributed systems, it is challenging to effectively and efficiently discover the anomalies from a voluminous amount of noisy and high-dimensional data streams. Moreover, the evolving of the system infrastructures brings new anomaly types that cannot be generalized as existing ones, making the existing anomaly detection solutions unavailable. To address these issues, in this paper, we introduce a new type of anomalies called contextual collective anomaly. Then we propose a framework to discover this type of anomaly over a collection of data streams in real time. A primary advantage of this solution is that it can accurately identify the anomalies by taking both the contextual information and the historical information of a data stream into consideration. Also, the proposed framework is designed in a way with a low computational cost, and is able to handle large-scale data streams. To demonstrate the effectiveness and efficiency of our proposed framework, we empirically validate it on a real world cluster.
    Full-text · Article · May 2015 · Expert Systems with Applications
  • Wei Tu · Tao Li · Qin Wang · Kun Cai · Xiang Gao · Hui Wang
    [Show abstract] [Hide abstract]
    ABSTRACT: The entire stx1 region from E. coli O157:H7, containing two open reading frames (stx1a and stx1b), was cloned into pET-32a with a single promoter. This region was transformed into E. coli TransB (DE3), which is a trxB and gor mutation strain. After expression in the E. coli periplasm in a completely soluble form, the rStx1 was purified and verified by SDS-PAGE, ELISA, and western blot analysis. Our rStx1 have Vero cell CD50 and LD50 values of approximately 30 ng and 1.5 μg, respectively. The final yield of the purified rStx1 ranged from 2 to 3 mg/L by one-step nickel affinity gel column chromatography. This method is an easy approach to the large-scale preparation of Stx1 at a reasonable cost. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
    No preview · Article · May 2015 · Biotechnology and Applied Biochemistry
  • [Show abstract] [Hide abstract]
    ABSTRACT: Microfluidic time gates are applied to control accurate flow time of liquids in the autonomous capillary systems, which are critical structures for point-of-care diagnostic and analytical applications. The time gate employs several abruptly changing geometry of the flow path to change the wettability of micro-channels to meet the requirement of delaying flow. However, obtaining desirable flow of liquids in microfluidics is still a limiting factor in the practical implementation. The work demonstrates a novel 2D serial cross-channel time gate and 3D serial cross-channel time gate. They are fabricated in PDMS-based autonomous capillary system. 2D serial cross-channel time gate is comprised of multiple paralleled channels of changeable width with dimensions from 300-800Îm. The number of the channels and the width variation of the cross intersections are crucial factors to influence the flow velocity of liquids. Compared with the 2D serial cross-channel time gate, the 3-D structures can eliminate the problem of entrapping air and improve the flow velocity of liquids in the time gate. The controlling time of the flow in 3D serial cross-channel time gate and 2D serial cross-channel time gate are 9~13s and 5s~51s, respectively.
    No preview · Article · May 2015 · Key Engineering Materials

  • No preview · Article · Apr 2015 · ACM Transactions on Intelligent Systems and Technology
  • Source
    Dataset: kdd12-tang
    Liang Tang · Tao Li · Larisa Shwartz

    Full-text · Dataset · Apr 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Personalized recommendation services have gained increasing popularity and attention in recent years as most useful information can be accessed online in real-time. Most on-line recommender systems try to address the information needs of users by virtue of both user and content information. Despite extensive recent advances, the problem of personalized recommendation remains challenging for at least two reasons. First, the user and item repositories undergo frequent changes, which makes traditional recommendation algorithms ineffective. Second, the so-called cold-start problem is difficult to address, as the information for learning a recommendation model is limited for new items. In this paper, we formulate personalized recommendation as a contextual bandit problem to balance the exploration/exploitation dilemma. Specifically in our work, we propose a parameter-free strategy for bandit problems, which employs a principled resampling approach called online boot- strap, to derive the sampling distribution of model parameter estimators. Under the paradigm of probability match- ing, the proposed algorithm randomly samples the model parameters to balance the tradeoff between exploration and exploitation. Extensive empirical experiments on two real- world collections of web data (including online advertising and news recommendation) demonstrate the effectiveness of our proposed approach in terms of the click-through rate. The experimental results also show that the performance of this parameter-free approach is robust in the cold-start situation, in which the analysts has no sufficient data or knowledge to tune parameters or choose a prior.
    Full-text · Conference Paper · Apr 2015

Publication Stats

7k Citations
428.90 Total Impact Points


  • 2015
    • Nanjing University of Posts and Telecommunications
      • Department of Computer Science & Technology
      Nan-ching, Jiangsu, China
  • 2013-2015
    • Lanzhou University Second Hospital
      Kao-lan-hsien, Gansu Sheng, China
    • Xiamen University
      • School of Information Science & Technology
      Amoy, Fujian, China
    • Lanzhou University
      • Institute of Meteorology
      Kao-lan-hsien, Gansu Sheng, China
    • Second Artillery Corps General Hospital
      Peping, Beijing, China
  • 2010-2015
    • Beijing Institute of Microbiology and Epidemiology
      Peping, Beijing, China
  • 2005-2015
    • Florida International University
      • School of Computing and Information Sciences
      Miami, Florida, United States
    • Sichuan University
      • • Department of Orthopedic Surgery
      • • College of Computer Science
      Hua-yang, Sichuan, China
  • 2004-2015
    • Sun Yat-Sen University
      • • State Key Laboratory of Oncology
      • • The First Affiliated Hospital
      Shengcheng, Guangdong, China
  • 2014
    • Dalian University of Technology
      Lü-ta-shih, Liaoning, China
  • 2009-2014
    • Fourth Military Medical University
      • Department of Orthodontics
      Xi’an, Liaoning, China
    • Academy of Military Medical Sciences
      T’ien-ching-shih, Tianjin Shi, China
  • 2007-2012
    • Sun Yat-Sen University of Medical Sciences
      • Reproductive Medical Center
      Shengcheng, Guangdong, China
  • 2011
    • Beijing University of Posts and Telecommunications
      Peping, Beijing, China
  • 2008-2009
    • Wuhan University
      Wu-han-shih, Hubei, China
    • Guangzhou Medical University
      Shengcheng, Guangdong, China
  • 2006
    • Washington International University
      International Falls, Minnesota, United States
  • 2001-2005
    • University of Rochester
      • Department of Computer Science
      Rochester, NY, United States
  • 2001-2002
    • Tsinghua University
      • • School of Materials Science and Engineering
      • • State Key Laboratory of New Ceramics and Fine Processing
      Peping, Beijing, China