28 Hits in 0.94 sec

Dual-attention Focused Module for Weakly Supervised Object Localization [article]

Yukun Zhou, Zailiang Chen, Hailan Shen, Qing Liu, Rongchang Zhao, and Yixiong Liang
2019 arXiv   pre-print
The research on recognizing the most discriminative regions provides referential information for weakly supervised object localization with only image-level annotations. However, the most discriminative regions usually conceal the other parts of the object, thereby impeding entire object recognition and localization. To tackle this problem, the Dual-attention Focused Module (DFM) is proposed to enhance object localization performance. Specifically, we present a dual attention module for
more » ... ion fusion, consisting of a position branch and a channel one. In each branch, the input feature map is deduced into an enhancement map and a mask map, thereby highlighting the most discriminative parts or hiding them. For the position mask map, we introduce a focused matrix to enhance it, which utilizes the principle that the pixels of an object are continuous. Between these two branches, the enhancement map is integrated with the mask map, aiming at partially compensating the lost information and diversifies the features. With the dual-attention module and focused matrix, the entire object region could be precisely recognized with implicit information. We demonstrate outperforming results of DFM in experiments. In particular, DFM achieves state-of-the-art performance in localization accuracy in ILSVRC 2016 and CUB-200-2011.
arXiv:1909.04813v1 fatcat:cpki345fbvdmnkdpzdnblnrhyu

Program Classification Using Gated Graph Attention Neural Network for Online Programming Service [article]

Mingming Lu, Dingwu Tan, Naixue Xiong, Zailiang Chen, Haifeng Li
2019 arXiv   pre-print
Chen are with the School of Computer Science and Engineering, Central South University, Changsha, China, 410083, e-mail: (,, N.  ... 
arXiv:1903.03804v1 fatcat:3kexjtwb4jeiznkgdiowijsytm

Hierarchical Contour Closure-Based Holistic Salient Object Detection

Qing Liu, Xiaopeng Hong, Beiji Zou, Jie Chen, Zailiang Chen, Guoying Zhao
2017 IEEE Transactions on Image Processing  
Zailiang Chen received the Ph.D. degree in computer science from Central South University in 2012.  ...  Chen was a cochair of International Workshops at ACCV2014, CVPR2016, and ICCV2017.  ... 
doi:10.1109/tip.2017.2703081 pmid:28500000 fatcat:rxkjth3jqraqjcatmrzizzm6pe

Saliency-Based Segmentation of Optic Disc in Retinal Images

Beiji Zou, Qing Liu, Kejuan Yue, Zailiang Chen, Jie Chen, Guoying Zhao
2019 Chinese journal of electronics  
Accurate segmentation of optic disc (OD) is significant for the automation of retinal analysis and retinal diseases screening. This paper proposes a novel optic disc segmentation method based on the saliency. It includes two stages:optic disc location and saliency-based segmentation. In the location stage, the OD is detected using a matched template and the density of the vessels. In the segmentation stage, we treat the OD as the salient object and formulate it as a saliency detection problem.
more » ... o measure the saliency of a region, the boundary prior and the connectivity prior are exploited. Then geodesic distance to the window boundary is computed to measure the cost the region spends to reach the window boundary. After a threshold and ellipse fitting, we obtain the OD. Experimental results on two public databases for OD segmentation show that the proposed method achieves thestate-of-the-art performance.
doi:10.1049/cje.2017.12.007 fatcat:jdcl5f7egraezovv73xy45s53i

Improved multi-scale line detection method for retinal blood vessel segmentation

Kejuan Yue, Beiji Zou, Zailiang Chen, Qing Liu
2018 IET Image Processing  
Changes of retinal blood vessel are precursors of many serious diseases such as diabetic retinopathy, hypertension and cardiovascular diseases. Automatic segmentation of retinal blood vessels in the fundus image can better assist in the diagnosis of these diseases and has been studied by many researchers. However, the segmentation of pale vessel pixels remains a problem because of their low contrasts with surrounding pixels. This study proposes an improved multi-scale line detector to segment
more » ... tinal vessels. It computes the line responses of vessels in multi-scale windows and takes the maximum as the response value, which can enhance the responses of pale vessel pixels near strong vessels or dark background pixels. Experimental results on the publicly available database DRIVE demonstrate that the proposed method can detect pale vessel pixels better. It achieves 75.28% in sensitivity and 94.47% in accuracy, which outperforms the state-of-the-art unsupervised methods. Compared with the supervised methods it also gets better sensitivity and comparable accuracy.
doi:10.1049/iet-ipr.2017.1071 fatcat:cvtpnsrzbzfkzicegfu6sixlxy

DDNet: Cartesian-polar Dual-domain Network for the Joint Optic Disc and Cup Segmentation [article]

Qing Liu and Xiaopeng Hong and Wei Ke and Zailiang Chen and Beiji Zou
2019 arXiv   pre-print
Existing joint optic disc and cup segmentation approaches are developed either in Cartesian or polar coordinate system. However, due to the subtle optic cup, the contextual information exploited from the single domain even by the prevailing CNNs is still insufficient. In this paper, we propose a novel segmentation approach, named Cartesian-polar dual-domain network (DDNet), which for the first time considers the complementary of the Cartesian domain and the polar domain. We propose a two-branch
more » ... of domain feature encoder and learn translation equivariant representations on rectilinear grid from Cartesian domain and rotation equivariant representations on polar grid from polar domain parallelly. To fuse the features on two different grids, we propose a dual-domain fusion module. This module builds the correspondence between two grids by the differentiable polar transform layer and learns the feature importance across two domains in element-wise to enhance the expressive capability. Finally, the decoder aggregates the fused features from low-level to high-level and makes dense predictions. We validate the state-of-the-art segmentation performances of our DDNet on the public dataset ORIGA. According to the segmentation masks, we estimate the commonly used clinical measure for glaucoma, i.e., the vertical cup-to-disc ratio. The low cup-to-disc ratio estimation error demonstrates the potential application in glaucoma screening.
arXiv:1904.08773v1 fatcat:hgxcnazyrfadzeeb6hqnhtams4

A Refined Equilibrium Generative Adversarial Network for Retinal Vessel Segmentation [article]

Yukun Zhou, Zailiang Chen, Hailan Shen, Xianxian Zheng, Rongchang Zhao, Xuanchu Duan
2019 arXiv   pre-print
Chen, X. Zheng, and R. Zhao are with the Department of Computer Science and Technology, Central South University, Changsha, Hunan, 410083 China. The corresponding author is H.  ... 
arXiv:1909.11936v2 fatcat:qkwnfwlwfvbpfb2tqswdsadf6u

A location-to-segmentation strategy for automatic exudate segmentation in colour retinal fundus images

Qing Liu, Beiji Zou, Jie Chen, Wei Ke, Kejuan Yue, Zailiang Chen, Guoying Zhao
2017 Computerized Medical Imaging and Graphics  
), (Wei Ke), (Kejuan Yue), (Zailiang Chen), (Guoying Zhao) the pixel level annotation is provided by an ophthalmologist.  ...  example for colour retinal fundus image with exudate regions and * I am corresponding author Email addresses: (Qing Liu), (Beiji Zou), (Jie Chen  ... 
doi:10.1016/j.compmedimag.2016.09.001 pmid:27665058 fatcat:kimdoqpjvvhctp47j5ppnlhbsm

Altered Spontaneous Brain Activity of Children with Unilateral Amblyopia: A Resting State fMRI Study

Peishan Dai, Jinlong Zhang, Jing Wu, Zailiang Chen, Beiji Zou, Ying Wu, Xin Wei, Manyi Xiao
2019 Neural Plasticity  
Objective. This study is aimed at investigating differences in local brain activity and functional connectivity (FC) between children with unilateral amblyopia and healthy controls (HCs) by using resting state functional magnetic resonance imaging (rs-fMRI). Methods. Local activity and FC analysis methods were used to explore the altered spontaneous brain activity of children with unilateral amblyopia. Local brain function analysis methods included the amplitude of low-frequency fluctuation
more » ... F). FC analysis methods consisted of the FC between the primary visual cortex (PVC-FC) and other brain regions and the FC network between regions of interest (ROIs-FC) selected by independent component analysis. Results. The ALFF in the bilateral frontal, temporal, and occipital lobes in the amblyopia group was lower than that in the HCs. The weakened PVC-FC was mainly concentrated in the frontal lobe and the angular gyrus. The ROIs-FC between the default mode network, salience network, and primary visual cortex network (PVCN) were significantly reduced, whereas the ROIs-FC between the PVCN and the high-level visual cortex network were significantly increased in amblyopia. Conclusions. Unilateral amblyopia may reduce local brain activity and FC in the dorsal and ventral visual pathways and affect the top-down attentional control. Amblyopia may also alter FC between brain functional networks. These findings may help understand the pathological mechanisms of children with amblyopia.
doi:10.1155/2019/3681430 pmid:31428144 pmcid:PMC6683781 fatcat:hyvvfguscrh5fe4pr65xmwfm7m

Region-segmentation strategy for Bruch's membrane opening detection in spectral domain optical coherence tomography images

Zailiang Chen, Peng Peng, Hailan Shen, Hao Wei, Pingbo Ouyang, Xuanchu Duan
2019 Biomedical Optics Express  
Bruch's membrane opening (BMO) is an important biomarker in the progression of glaucoma. Bruch's membrane opening minimum rim width (BMO-MRW), cup-to-disc ratio in spectral domain optical coherence tomography (SD-OCT) and lamina cribrosa depth based on BMO are important measurable parameters for glaucoma diagnosis. The accuracy of measuring these parameters is significantly affected by BMO detection. In this paper, we propose a method for automatically detecting BMO in SD-OCT volumes accurately
more » ... to reduce the impact of the border tissue and vessel shadows. The method includes three stages: a coarse detection stage composed by retinal pigment epithelium layer segmentation, optic disc segmentation, and multi-modal registration; a fixed detection stage based on the U-net in which BMO detection is transformed into a region segmentation problem and an area bias component is proposed in the loss function; and a post-processing stage based on the consistency of results to remove outliers. Experimental results show that the proposed method outperforms previous methods and achieves a mean error of 42.38 μm.
doi:10.1364/boe.10.000526 pmid:30800497 pmcid:PMC6377878 fatcat:lg6pi6r4ongzfiez6fvuyaalha

Downregulation of microRNA-196a enhances the sensitivity of non-small cell lung cancer cells to cisplatin treatment

2016 International Journal of Molecular Medicine  
MicroRNAs (miRNAs or miRs) are a class of small, non-coding RNA molecules that play an important role in the pathogenesis of human diseases through the regulation of gene expression. Although miRNA-196a has been implicated in the progression of human lung cancer, its role in enhancing the sensitivity of non-small cell lung cancer (NSCLC) cells to cisplatin has not yet been confirmed. The aim of this study was to evaluate the effects of miRNA-196a on the sensitivity of NSCLC cells to cisplatin
more » ... vitro and in vivo. RT-qPCR was used to detect miRNA-196a expression. Synthesized locked nucleic acid (LNA)-anti-miRNA-196a oligonucleotide was transiently transfected into the SPC-A-1 and A549 lung cancer cells to examine the effects of miRNA-196a on the growth of and colony formation inthe cisplatin-treated cells. The effects of miRNA-196a on the sensitivity of SPC-A-1 cells to cisplatin in vivo were determined using BALB/c nude mice. The expression of miRNA-196a was significantly higher in both the lung cancer tissues and cell lines. The LNA-based knockdown of miRNA-196a significantly inhibited SPC-A-1 and A549 cell growth and induced apoptosis. Moreover, the downregulation of miRNA-196a sensitized the SPC-A-1 and A549 NSCLC cells to cisplatin in vitro and in vivo, by inducing apoptosis. The findings of this study demonstrate that the administration of cisplatin in combination with miRNA-196a-targeted therapy may be a potential therapeutic strategy for the treatment of NSCLC.
doi:10.3892/ijmm.2016.2513 pmid:26936095 fatcat:srljtbx25rhzlpj6yfvuxtwzh4

Infusion of Trx-1-Overexpressing hucMSC Prolongs the Survival of Acutely Irradiated NOD/SCID Mice by Decreasing Excessive Inflammatory Injury

JiangWei Hu, ZaiLiang Yang, Jun Wang, YongYong Tang, Hao Liu, Bin Zhang, Hu Chen, Eva Mezey
2013 PLoS ONE  
A protective reagent for ARI should have the ability to repair injured tissue caused by radiation and prevent continuous damage from secondary risk factors. Trx-1 was explored as a candidate therapy for ARI, as it scavenges reactive oxygen species, regulates cell growth and differentiation, participates in immune reactions, and inhibits apoptosis by acting inside and/or outside cells. Trx-1 can also decrease excessive inflammation in ARI by regulating the creation of inflamed media, by
more » ... g the activation of complement, and by reducing the chemotaxis, adhesion, and migration of inflammatory cells. As effectively and stably expressing exogenous genes in the long term and regulating immune inflammation and tissue repair, MSC are a good choice for Trx-1 gene therapy. In this study, Trx-1-overexpressing hucMSC-Trx-1 were obtained by adenoviral vector-mediated infection. We first measured the redox capacity of hucMSC-Trx-1 with an antioxidant capacity (T-AOC) assay, a hydrogen peroxide (H 2 O 2 ) content determination assay in vivo, a H 2 O 2 -induced oxidation hemolysis assay, and a lipid peroxidation assay in vitro. Then, we measured survival time, the protection of the hematopoietic system, and the regulation of inflammation in important organs in three treatment groups of NOD/SCID mice (treated with hucMSC-Trx-1, with hucMSC, and with saline) that were exposed to 4.5 Gy 60 Co-c-ray radiation. The hucMSC-Trx-1 group achieved superior antioxidation results, protecting bone marrow hematopoietic stem cells (Lin 2 CD117 + : hucMSC-Trx-1 vs. hucMSC, P,0.05; hucMSC-Trx-1 vs. NS, P,0.01), promoting the formation of red blood cells and hemoglobin (hucMSC-Trx-1 vs. hucMSC or NS, P,0.05), reducing inflammation and damage in important organs (Bone marrow and lung: hucMSC-Trx-1 vs. NS, P,0.01; hucMSC-Trx-1 vs. hucMSC, P,0.05. Liver and intestine: hucMSC-Trx-1 vs. NS, P,0.05; hucMSC-Trx-1 vs. hucMSC, P,0.05), and prolonging survival (hucMSC-Trx-1 vs. hucMSC or NS, P,0.01). Therefore, hucMSC-Trx-1 combines the merits of gene and cell therapy as a multifunctional radioprotector for ARI.
doi:10.1371/journal.pone.0078227 pmid:24223778 pmcid:PMC3817237 fatcat:yg6rshujefgmzayanycn6mfn64

Altered Effective Connectivity of Children and Young Adults With Unilateral Amblyopia: A Resting-State Functional Magnetic Resonance Imaging Study

Peishan Dai, Xiaoyan Zhou, Yilin Ou, Tong Xiong, Jinlong Zhang, Zailiang Chen, Beiji Zou, Xin Wei, Ying Wu, Manyi Xiao
2021 Frontiers in Neuroscience  
., 2017; Chen et al., 2018; Ma et al., 2018) . EC can be analyzed by various methods.  ...  Copyright © 2021 Dai, Zhou, Ou, Xiong, Zhang, Chen, Zou, Wei, Wu and Xiao. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY).  ... 
doi:10.3389/fnins.2021.657576 fatcat:ve477gv7mvcxjfbkixe4nz5rka

Application of Improved U-Net Convolutional Neural Network for Automatic Quantification of the Foveal Avascular Zone in Diabetic Macular Ischemia

Yongan Meng, Hailei Lan, Yuqian Hu, Zailiang Chen, Pingbo Ouyang, Jing Luo, Ali Sharif
2022 Journal of Diabetes Research  
Objectives.The foveal avascular zone (FAZ) is a biomarker for quantifying diabetic macular ischemia (DMI), to automate the identification and quantification of the FAZ in DMI, using an improved U-Net convolutional neural network (CNN) and to establish a CNN model based on optical coherence tomography angiography (OCTA) images for the same purpose. Methods. The FAZ boundaries on the full-thickness retina of 6 × 6 mm en face OCTA images of DMI and normal eyes were manually marked. Seventy percent
more » ... of OCTA images were used as the training set, and ten percent of these images were used as the validation set to train the improved U-Net CNN with two attention modules. Finally, twenty percent of the OCTA images were used as the test set to evaluate the accuracy of this model relative to that of the baseline U-Net model. This model was then applied to the public data set sFAZ to compare its effectiveness with existing models at identifying and quantifying the FAZ area. Results. This study included 110 OCTA images. The Dice score of the FAZ area predicted by the proposed method was 0.949, the Jaccard index was 0.912, and the area correlation coefficient was 0.996. The corresponding values for the baseline U-Net were 0.940, 0.898, and 0.995, respectively, and those based on the description data set sFAZ were 0.983, 0.968, and 0.950, respectively, which were better than those previously reported based on this data set. Conclusions. The improved U-Net CNN was more accurate at automatically measuring the FAZ area on the OCTA images than the traditional CNN. The present model may measure the DMI index more accurately, thereby assisting in the diagnosis and prognosis of retinal vascular diseases such as diabetic retinopathy.
doi:10.1155/2022/4612554 pmid:35257013 pmcid:PMC8898103 fatcat:g7vfqadhmjdefkeynovivubwyy

Weakly-Supervised Simultaneous Evidence Identification and Segmentation for Automated Glaucoma Diagnosis

Rongchang Zhao, Wangmin Liao, Beiji Zou, Zailiang Chen, Shuo Li
Although most previous works are devoted to segmentation of optic disc as evident for glaucoma diagnosis (Cheng et al. 2013; Chen et al. 2015; Almazroa et al. 2015) , designing the well-performing model  ...  MTL) is a learning paradigm in machine learning and its aim is to leverage shared information contained in multiple related tasks to help improve the performance of all the tasks (Zhang and Yang 2017; Chen  ... 
doi:10.1609/aaai.v33i01.3301809 fatcat:ew2gyjxrrve5bip7hm42dgoapu
« Previous Showing results 1 — 15 out of 28 results