Article
Computer Science, Artificial Intelligence
Lei Zhu, Xuejing Kang, Lizhu Ye, Anlong Ming
Summary: This paper proposes an ENCUT model that establishes a balanced graph model by adopting a meaningful-loop and a k-step random walk to enhance small object segmentation. The model is further improved by adding a new RWRT that adds local attention to the segmentation of twigs. Experimental results show that the model achieves state-of-the-art performance among NCut-based segmentation models.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Yingjie Yin, De Xu, Xingang Wang, Lei Zhang
Summary: The proposed DDEAL method for fast VOS does not rely on online fine-tuning and achieves state-of-the-art performance on two datasets with fast speed and minimal accuracy loss in a faster version.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Hu Lu
Summary: This study proposes a new interactive segmentation framework that allows users to cut an object from the background with just one click. By combining automatic segmentation and user interaction, the framework achieves image interactive segmentation successfully.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Mehrnaz Niazi, Kambiz Rahbar, Mansour Sheikhan, Maryam Khademi
Summary: This paper investigates entropy-based kernel graph cut image segmentation and proposes a method that incorporates a 2-layer feature space to improve segmentation performance. The proposed method is particularly effective in dealing with non-textural and complex textural images. Experimental results demonstrate the superior performance of the proposed method in energy-based image segmentation.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Shaojun Qu, Huang Tan, Qiaoliang Li, Zili Peng
Summary: The paper proposes a graph-based image segmentation model and improves its performance by adding spatial distance and contour orientation energy terms and modifying the construction of the energy graph. Experimental results show that the proposed method outperforms many other methods on multiple datasets.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2022)
Article
Engineering, Biomedical
Katarzyna Hajdowska, Sebastian Student, Damian Borys
Summary: The GRABaCELL algorithm combines Graph Cut, Watershed segmentation, and Hough Circular Transform to improve automatic segmentation and counting of living cells. The introduced modified accuracy metric helps assess the quality of segmentation based on the number of cells detected in the image. The results of the GRABaCELL method outperform other compared methods in visual assessment, with both Dice index and modified accuracy metric showing significantly better performance.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Information Systems
Jiapei Feng, Xinggang Wang, Wenyu Liu
Summary: This paper proposes a weakly-supervised task for pixel-level semantic segmentation using semi-supervised learning methods, achieving fine segmentation from image-level labels. Combining the graph cut algorithm with activation seeds generated by a classification network provides effective supervision information, leading to successful experimental results.
SCIENCE CHINA-INFORMATION SCIENCES
(2021)
Article
Robotics
Minho Oh, Euigon Jung, Hyungtae Lim, Wonho Song, Sumin Hu, Eungchang Mason Lee, Junghee Park, Jaekyung Kim, Jangwoo Lee, Hyun Myung
Summary: This paper presents a method that performs simultaneous traversable ground detection and object clustering using a graph representation of a 3D point cloud. By leveraging the structure of nodes and edges, real-time operation is achieved and over-segmentation is mitigated. Experimental results demonstrate that the proposed method outperforms other state-of-the-art methods in terms of traversable ground segmentation.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Engineering, Biomedical
Hongming Xu, Lina Liu, Xiujuan Lei, Mrinal Mandal, Cheng Lu
Summary: This study introduces a new unsupervised method, TisCut, for assisting tissue image segmentation and annotations, showing comparative performance with U-Net models in necrosis and melanoma detections.
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS
(2021)
Article
Computer Science, Artificial Intelligence
Petr Taborsky, Laurent Vermue, Maciej Korzepa, Morten Morup
Summary: The article introduces a novel Bayesian probabilistic model for graph cutting, providing an effective solution to separating community structures in complex networks. The method demonstrates excellent performance on real social networks and image segmentation problems, while also learning the parameter space.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Software Engineering
Yadang Chen, Duolin Wang, Zhiguo Chen, Zhi-Xin Yang, Enhua Wu
Summary: We propose a lightweight and efficient semi-supervised video object segmentation network based on the space-time memory framework. Our model achieves high-performance, real-time segmentation by integrating multi-frame image information without increased memory usage. The spatial constraint module effectively alleviates mismatching of similar targets. Our model achieves state-of-the-art results on various datasets.
COMPUTATIONAL VISUAL MEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Minh On Vu Ngoc, Edwin Carlinet, Jonathan Fabrizio, Thierry Geraud
Summary: This paper presents a method that combines the Dahu pseudo-distance with edge information in a graph-cut optimization framework, leveraging their complementary strengths. The method achieves better performance in noisy and blurred images compared to other distance-based and graph-cut methods, reducing user effort in object selection.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Alessandro Gnutti, Fabrizio Guerrini, Riccardo Leonardi
Summary: This work addresses the challenging problem of reflection symmetry detection in unconstrained environments by proposing a two-stage solution, which establishes a better correspondence between the outcomes of the algorithm and a human-constructed ground truth, achieves significant performance gains compared to recent symmetry detection competitions, and further validates the approach through perceptual validation experiments with users on a newly built dataset.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Hui Wang, Weibin Liu, Weiwei Xing
Summary: More and more researchers are focusing on video object segmentation as it is crucial for various computer vision applications, however, challenges such as appearance changes and background distractions persist. This paper introduces a novel neural network that addresses these challenges efficiently.
APPLIED INTELLIGENCE
(2022)
Article
Automation & Control Systems
Zheyun Qin, Xiankai Lu, Xiushan Nie, Dongfang Liu, Yilong Yin, Wenguan Wang
Summary: We propose a novel method using a new generative model to automatically detect, segment, and track instances in a video sequence. Our hierarchical structural embedding learning predicts high-quality masks with accurate boundary details using normalizing flows. The method achieves superior performance on video instance segmentation benchmarks and demonstrates generalizability on an unsupervised video object segmentation dataset.
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2023)
Article
Biochemical Research Methods
Xiaohan Xing, Fan Yang, Hang Li, Jun Zhang, Yu Zhao, Mingxuan Gao, Junzhou Huang, Jianhua Yao
Summary: This study proposes a novel multi-level attention graph neural network (MLA-GNN) for disease diagnosis and prognosis, which investigates gene association information and co-functional gene modules to facilitate disease state prediction. The experimental results demonstrate that MLA-GNN achieves state-of-the-art performance on transcriptomic and proteomic data, and the selected relevant genes are consistent with clinical understanding.
Article
Computer Science, Interdisciplinary Applications
Jiahong Wei, Guijie Zhu, Zhun Fan, Jinchao Liu, Yibiao Rong, Jiajie Mo, Wenji Li, Xinjian Chen
Summary: This study proposes a novel automated design method (Genetic U-Net) that generates a U-shaped convolutional neural network for better retinal vessel segmentation with fewer parameters. Experimental results show that the proposed method outperforms other state-of-the-art models and has significantly fewer parameters.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Computer Science, Interdisciplinary Applications
Meng Wang, Weifang Zhu, Fei Shi, Jinzhu Su, Haoyu Chen, Kai Yu, Yi Zhou, Yuanyuan Peng, Zhongyue Chen, Xinjian Chen
Summary: In this paper, a novel multi-scale transformer global attention network (MsTGANet) is proposed for drusen segmentation in retinal OCT images. The network utilizes a multi-scale transformer non-local module and a multi-semantic global channel and spatial joint attention module to improve segmentation accuracy.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Clinical Neurology
Chang Jianbo, Pei Hanqi, Chen Yihao, Jiang Cheng, Shang Hong, Wang Yuxiang, Wang Xiaoning, Ye Zeju, Wang Xingong, Tian Fengxuan, Chai Jianjun, Xu Jijun, Li Zhaojian, Ma Wenbin, Wei Junji, Jianhua Yao, Feng Ming, Wang Renzhi
Summary: An artificial intelligence model utilizing weakly supervised multitask learning structure was able to identify onset time for spontaneous intracerebral hemorrhage patients with unclear onset, potentially benefiting from time-dependent treatments. The model showed good performance and generalizability, demonstrating potential for integration into clinical practice.
INTERNATIONAL JOURNAL OF STROKE
(2022)
Article
Computer Science, Information Systems
Changqing Yang, Xinxin Zhou, Weifang Zhu, Dehui Xiang, Zhongyue Chen, Jin Yuan, Xinjian Chen, Fei Shi
Summary: This paper aims to develop an automatic method for nerve fiber segmentation from in vivo corneal confocal microscopy (CCM) images. A novel multi-discriminator adversarial convolutional network (MDACN) framework is proposed, and experimental results show that the method has excellent segmentation performance for corneal nerve fibers.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Yuhe Shen, Jiang Li, Weifang Zhu, Kai Yu, Meng Wang, Yuanyuan Peng, Yi Zhou, Liling Guan, Xinjian Chen
Summary: In this paper, a novel approach called graph attention U-Net (GA-UNet) is proposed for retinal layer surface detection and CNV segmentation in OCT images. The method addresses the challenge of retinal layer deformation caused by CNV and achieves improved performance through the use of graph attention encoder and graph decorrelation module. The proposed approach also introduces a new loss function to preserve the correct topological order and boundary continuity of retinal layers. Experimental results demonstrate that the proposed model outperforms existing methods and achieves new state of the art on the datasets.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Multidisciplinary Sciences
Rongbo Shen, Lin Liu, Zihan Wu, Ying Zhang, Zhiyuan Yuan, Junfu Guo, Fan Yang, Chao Zhang, Bichao Chen, Wanwan Feng, Chao Liu, Jing Guo, Guozhen Fan, Yong Zhang, Yuxiang Li, Xun Xu, Jianhua Yao
Summary: Spatial-ID is a supervision-based cell typing method that combines the existing knowledge of reference single-cell RNA-seq data and the spatial information of spatially resolved transcriptomics data. Benchmarking analyses on publicly available datasets and application on self-collected mouse brain hemisphere dataset validate the superiority and scalability of Spatial-ID.
NATURE COMMUNICATIONS
(2022)
Article
Engineering, Biomedical
Lingjiao Pan, Xinjian Chen
Summary: This paper presents retinal OCT image registration methods and their clinical applications. The paper systematically reviews registration methods based on volumetric transformation and image features. Furthermore, the applications of these methods in correcting scanning artifacts, reducing speckle noise, fusing and splicing images, and evaluating disease progression are studied. The paper also discusses the registration of retina with serious pathology and registration with deep learning technique.
IEEE REVIEWS IN BIOMEDICAL ENGINEERING
(2023)
Article
Computer Science, Interdisciplinary Applications
Jianye Pang, Cheng Jiang, Yihao Chen, Jianbo Chang, Ming Feng, Renzhi Wang, Jianhua Yao
Summary: Combining vision transformer with CNN in medical volume dense prediction shows promise and challenges. This paper proposes a novel 3D Shuffle-Mixer network using a local vision transformer-MLP paradigm for improved dense prediction in medical images. Experimental results demonstrate the superiority of the proposed model compared to other state-of-the-art methods for medical dense prediction.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Computer Science, Interdisciplinary Applications
Dehui Xiang, Shenshen Yan, Ying Guan, Mulin Cai, Zheqing Li, Haiyun Liu, Xinjian Chen, Bei Tian
Summary: In this paper, a new segmentation strategy called a dual stream segmentation network embedded into a conditional generative adversarial network is proposed to improve the accuracy of retinal lesion segmentation. The proposed method is cross-validated in 384 clinical fundus fluorescein angiography images and 1040 optical coherence tomography images. Compared to state-of-the-art methods, the proposed method can achieve better segmentation of retinal capillary non-perfusion region and choroidal neovascularization.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Engineering, Biomedical
Ivica Kopriva, Fei Shi, Mingyig Lai, Marija Stanfel, Haoyu Chen, Xinijan Chen
Summary: This article presents a method for compressing and de-speckling 3D optical coherence tomography (OCT) images using low-rank tensor approximations and non-convex optimization problems. The results show that the proposed method can generate higher quality compressed images and is applicable for machine learning-based and visual inspection-based diagnosis.
PHYSICS IN MEDICINE AND BIOLOGY
(2023)
Article
Computer Science, Information Systems
Pengfei Zhang, Xinjian Chen, Ziting Yin, Xin Zhou, Qingxin Jiang, Weifang Zhu, Dehui Xiang, Yun Tang, Fei Shi
Summary: In this paper, a novel feature augment network (FANet) is proposed for automatic segmentation of skin wounds, and an interactive feature augment network (IFANet) is designed to provide interactive adjustment on the automatic segmentation results. The results indicate that the FANet gives good segmentation results while the IFANet can effectively improve them based on simple marking.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2023)
Meeting Abstract
Medicine, Research & Experimental
Xiaoying Lou, Niyun Zhou, Lili Feng, Zhenhui Li, Yuqi Fang, Xinjuan Fan, Hailing Liu, Jianhua Yao, Yan Huang
LABORATORY INVESTIGATION
(2022)
Meeting Abstract
Pathology
Hailing Liu, Yu Zhao, Fan Yang, Xiaoying Lou, Jianhua Yao, Xinjuan Fan
Meeting Abstract
Pathology
Xiaoying Lou, Niyun Zhou, Lili Feng, Zhenhui Li, Yuqi Fang, Xinjuan Fan, Hailing Liu, Jianhua Yao, Yan Huang
