检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Liu, Shaolei Wang, Manning Song, Zhijian Digital Medical Research Center School of Basic Medical Science Fudan University China Shanghai Key Laboratory of Medical Image Computing and Computer Assited Intervention

We propose an unsupervised image fusion architecture for multiple application scenarios based on the combination of multi-scale discrete wavelet transform through regional energy and deep learning. To our best knowledge, this is the first time that a conventional frequency method has been combined with deep learning for feature maps fusion. The useful information of feature maps can be utilized adequately through multi-scale discrete wavelet transform in our proposed method. Compared with other state-of-the-art fusion methods, the proposed algorithm exhibits better fusion performance in both subjective and objective evaluation. Moreover, it's worth mentioning that comparable fusion performance trained in COCO dataset can be obtained by training with a much smaller dataset with only hundreds of images chosen randomly from COCO. Hence, the training time is shortened substantially, leading to the improvement of the model's performance both in practicality and training efficiency. Copyright © 2020, The Authors. All rights reserved.

关键词： image fusion

来源：评论

学校读者我要写书评

暂无评论

An entropy weighted nonnegative matrix factorization algorithm for feature representation

arXiv

引用

arXiv 2021年

作者： Wei, Jiao Tong, Can Wu, Bingxue He, Qiang Qi, Shouliang Yao, Yudong Teng, Yueyang College of Medicine and Biological Information Engineering Northeastern University Shenyang110169 China Department of Electrical and Computer Engineering Stevens Institute of Technology HobokenNJ07030 United States College of Medicine and Biological Information Engineering Northeastern University Shenyang110169 China Key Laboratory of Intelligent Computing in Medical Image Ministry of Education Shenyang110169 China

Nonnegative matrix factorization (NMF) has been widely used to learn low-dimensional representations of data. However, NMF pays the same attention to all attributes of a data point, which inevitably leads to inaccurate representation. For example, in a human-face data set, if an image contains a hat on the head, the hat should be removed or the importance of its corresponding attributes should be decreased during matrix factorizing. This paper proposes a new type of NMF called entropy weighted NMF (EWNMF), which uses an optimizable weight for each attribute of each data point to emphasize their importance. This process is achieved by adding an entropy regularizer to the cost function and then using the Lagrange multiplier method to solve the problem. Experimental results with several data sets demonstrate the feasibility and effectiveness of the proposed method. We make our code available at https://***/Poisson-EM/ Entropy-weighted-NMF. © 2021, CC BY.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Development of a Nomogram for the Identification of Meniscus Injury by Subjective Symptoms

引用

Forest Chemicals Review 2021年第NOVEMBER-DECEMBER期2021卷 408-420页

作者： Wang, Guangjun Zhang, Hanyuan Wang, Yu Chen, Ying Sun, Yining Wang, Shijun Su, Benyue Ma, Zuchang Anhui Province Key Laboratory of Medical Physics and Technology Institute of Intelligent Machines Hefei Institutes of Physical Science Chinese Academy of Sciences Hefei230031 China Science Island Branch of Graduate School University of Science and Technology of China Hefei230026 China The University Key Laboratory of Intelligent Perception and Computing of Anhui Province Anqing Normal University Anqing246013 China Department of Sports Medicine and Arthroscopic Surgery The First Affiliated Hospital of Anhui Medical University Hefei230022 China Dalian medical university Dalian116044 China

Background: Our study aimed to construct a nomogram with a scoring system based on subjective symptoms to identify meniscus injury. Methods: This study recruited 157 participants for a cross-sectional study. The doctor used MRI to diagnose each participant's knee joint. We use questionnaires to collect data on 14 subjective symptoms of each patient. Chi-square test and logistic regression were used for statistical analysis to screen for significant symptoms of meniscus injury relative to other knee diseases. We used the nomogram method to score the significant symptoms and build a scoring model. Results: Multivariate analysis showed that Pain Activity (OR = 3.41), Pain Hyperflexion (OR = 4.135), Tend Knee Space (OR = 62.138) were statistically significant risk symptoms for meniscus injury. Knee dislocation (OR = 0.184) was not a significant distinguishing symptom. Analysis of the nomogram model showed that the total score for each symptom ranged from 37 to 219, with corresponding risk rates ranging from 0.10 to 0.95 points. The C-index was calculated to assess the recognition accuracy of this nomogram scoring system was 88.75% (95% CI 85.24%-90.78%). Conclusions: We found that using the nomogram to establish an identification model to distinguish meniscus injury from other knee diseases based on subjective symptoms was effective. This method is a convenient and effective tool to evaluate meniscus injury and support the prevention and self-management of meniscus injury. © 2021 Kriedt Enterprises Ltd. All Rights Reserved.

关键词： Surveys

来源：评论

学校读者我要写书评

暂无评论

EndoBoost: a plug-and-play module for false positive suppression during computer-aided polyp detection in real-world colonoscopy (with dataset)

arXiv

引用

arXiv 2022年

作者： Wang, Haoran Zhu, Yan Qin, Wenzheng Zhang, Yizhe Zhou, Pinghong Li, Quanlin Wang, Shuo Song, Zhijian Digital Medical Research Center School of Basic Medical Sciences Fudan University Shanghai200032 China Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention Shanghai200032 China Endoscopy Center Endoscopy Research Institute Zhongshan Hospital Fudan University Shanghai200032 China Shanghai Collaborative Innovation Center of Endoscopy Shanghai200032 China School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu210014 China

The advance of computer-aided detection systems using deep learning opened a new scope in endoscopic image analysis. However, the learning-based models developed on closed datasets are susceptible to unknown anomalies in complex clinical environments. In particular, the high false positive rate of polyp detection remains a major challenge in clinical practice. In this work, we release the FPPD-13 dataset, which provides a taxonomy and real-world cases of typical false positives during computer-aided polyp detection in real-world colonoscopy. We further propose a post-hoc module EndoBoost, which can be plugged into generic polyp detection models to filter out false positive predictions. This is realized by generative learning of the polyp manifold with normalizing flows and rejecting false positives through density estimation. Compared to supervised classification, this anomaly detection paradigm achieves better data efficiency and robustness in open-world settings. Extensive experiments demonstrate a promising false positive suppression in both retrospective and prospective validation. In addition, the released dataset can be used to perform ‘stress’ tests on established detection systems and encourages further research toward robust and reliable computer-aided endoscopic image analysis. The dataset and code will be publicly available at http://***. © 2022, CC BY.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

An Automatic Extraction Method of Coronary Arteries Centerline Based on Cellular Automata

An Automatic Extraction Method of Coronary Arteries Centerli...

引用

作者： Guangkun Ma Jinzhu Yang Hong Zhao School of Computer Science and Engineering Northeastern University School of Software Shenyang University of Technology Key Laboratory of Intelligent Computing in Medical Image Ministry of Education School of Computer Science and Engineering Northeastern University

Effectively obtaining the centerline of the coronary arteries is very important to the clinical medical diagnosis. Cellular automata are extensively used to find fine solutions in many complex hard. However, it has the problem of large amount of computation. In this paper, we propose an automatic extraction method of coronary centerline based on cellular automata. Different from the existing methods, the distance from boundary field is used to extract the centerline of the coronary arteries roughly, which can place the candidate voxel at the center of the coronary arteries and reduce the size of the data. Furthermore, the centerlines of the coronary arteries are extracted accurately by using cellular automata to find the shortest path. The proposed method can extract the centerlines of each branch of the coronary arteries simultaneously from a single source to multiple destinations. Furthermore, compared to the method in Ref [10], our proposed method works well, especially in the case of large vessel curvature, the voxels at the extracted centerline are relatively centered. In the last, quantitative evaluation is conducted on the coronary arteries and the results show that the proposed method achieves a smaller deviation distance.

关键词： Cellular automata Distance from Boundary Coronary arteries Computerized Tomography Angiography

来源：评论

学校读者我要写书评

暂无评论

Application of machine learning on the modelling of diffusion magnetic resonance imaging signal

Application of machine learning on the modelling of diffusio...

引用

2020 International Seminar on Artificial Intelligence, Networking and Information Technology, AINIT 2020

作者： Ye, Chen Xu, Daoyun Wang, Lihui Zhu, Yuemin Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province School of Computer Science and Technology Guizhou University Guiyang China CNRS INSERM CREATIS UMR 5220 U1206 F-69621 Univ Lyon INSA Lyon Lyon France

The modelling of diffusion Magnetic Resonance Imaging (dMRI) signals is very important for medical clinical application. However, the traditional method is to use a fixed mathematical model to make assumptions about the diffusion-weighted (DW) signals of all regions of human organ, which is unreasonable. In this paper, Convolutional Neural Network (CNN), a machine learning based method is used for learning the different characteristics of the signals, and finally intelligently give multi-model predictions for different regions of human livers. The performance of the proposed method is verified on both simulation and real liver data. The results show that the multi-model predicted by CNN method has high performance in distinguishing normal liver from diseased liver, and has great clinical application prospects © Published under licence by IOP Publishing Ltd.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

A Strategy for Identifying the Risk of Developing Alzheimer's Disease from Healthy Controls: Distance to Novelty Detection Boundary

SSRN

引用

SSRN 2022年

作者： Mao, Jiangbing Ye, Qinyong Yang, Hongqin Bucholc, Magda Liu, Shuo Gao, Wenzhao Pan, Jie Xin, Jiawei Ding, Xuemei Fujian Provincial Key Laboratory for Photonics Technology Key Laboratory of OptoElectronic Science and Technology for Medicine of Ministry of Education Fujian Normal University Fuzhou350007 China Department of Neurology Fujian Medical University Union Hospital Fuzhou350001 China School of Computing Engineering and Intelligent Systems Ulster University DerryNI LondonderryBT48 7JL United Kingdom Xiamen Jingyi Zhikang Medical Technology Co. Ltd. No. 59 Huli Avenue Huli District Xiamen361000 China

This study first employed four interpretable novelty detection (ND) algorithms to detect the severity of cognitive disorder of individuals. The detector was modeled based on the healthy controls (HC) data only. Datasets used are respectively from the Australian Imaging Biomarkers and Lifestyle Study of Ageing project and the Fujian medical University Union Hospital, China. A DtB (distance to ND boundary) strategy was proposed to visually detect the severity of cognitive decline of potential mild cognitive impairment (MCI) and Alzheimer’s disease (AD) from HC, in an early stage. With non-invasive and cost-effective cognitive and functional assessments modality, the best overall performance of detecting MCI&AD for the AIBL and FMUUH was obtained by using Mixture of Gaussian-based ND algorithm, with AUC of 0.8757 and 0.9443, respectively. More practically, an interactive web-based graphical user interface tool was further developed for non-technical stakeholders to easily utilize our proposed strategy for supporting MCI/AD diagnosis. © 2022, The Authors. All rights reserved.

关键词： Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

Systematic Study of Joint Influence of Angular Resolution and Noise in Cardiac Diffusion Tensor Imaging 11th

Systematic Study of Joint Influence of Angular Resolution an...

引用

11th International Conference on Functional Imaging and Modeling of the Heart, FIMH 2021

作者： He, Yunlong Wang, Lihui Yang, Feng Xia, Yong Clarysse, Patrick Zhu, Yuemin Univ Lyon INSA Lyon Université Claude Bernard Lyon 1 CNRS Inserm CREATIS UMR 5220 U1294 Lyon69621 France Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province School of Computer Science and Technology Guizhou University Guiyang China National Library of Medicine National Institute of Health 8600 Rockvill Pike BethesdaMD20894 United States School of Computer Science NPU Xian China

ISBN: (纸本)9783030787097

Diffusion tensor imaging (DTI) is a promising imaging technique to non-invasively study diffusion properties and fiber structures of myocardial tissues. Previous studies have investigated the influence of noise or angular resolution independently on the estimation of diffusion tensors in DTI. However, the joint influence of these two factors in DTI remains unclear. In this paper, we propose to systematically study the joint influence of angular resolutions and noise levels on the estimation of diffusion tensors and tensor-derived fractional anisotropy (FA) and mean diffusivity (MD). The results showed that, as expected, given a certain noise level and sufficient acquisition time, the accuracy of diffusion tensor, FA and MD all increase as the angular resolution. Moreover, when the angular resolution reached a certain value, further increasing the number of angular resolutions has little effect on the estimation of diffusion tensor, FA and MD. Also, both the mean and variance of FA or MD decrease as the angular resolution increases. For an imposed acquisition time, increasing the angular resolution reduces SNR of DW images. When fixing SNR, higher angular resolution can be obtained at the expense of longer acquisition time. These findings suggest the necessity of an optimized trade-off when designing DTI protocols. © 2021, Springer Nature Switzerland AG.

关键词： Diffusion tensor imaging

来源：评论

学校读者我要写书评

暂无评论

Learning bronchiole-sensitive airway segmentation cnns by feature recalibration and attention distillation 23rd

Learning bronchiole-sensitive airway segmentation cnns by fe...

引用

23rd International Conference on medical image computing and Computer-Assisted Intervention, MICCAI 2020

作者： Qin, Yulei Zheng, Hao Gu, Yun Huang, Xiaolin Yang, Jie Wang, Lihui Zhu, Yue-Min Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province College of Computer Science and Technology Guizhou University Guiyang China UdL INSA Lyon CREATIS CNRS UMR 5220 INSERM U1206 Lyon France

ISBN: (纸本)9783030597092

Training deep convolutional neural networks (CNNs) for airway segmentation is challenging due to the sparse supervisory signals caused by severe class imbalance between long, thin airways and background. In view of the intricate pattern of tree-like airways, the segmentation model should pay extra attention to the morphology and distribution characteristics of airways. We propose a CNNs-based airway segmentation method that enjoys superior sensitivity to tenuous peripheral bronchioles. We first present a feature recalibration module to make the best use of learned features. Spatial information of features is properly integrated to retain relative priority of activated regions, which benefits the subsequent channel-wise recalibration. Then, attention distillation module is introduced to reinforce the airway-specific representation learning. High-resolution attention maps with fine airway details are passing down from late layers to previous layers iteratively to enrich context knowledge. Extensive experiments demonstrate considerable performance gain brought by the two proposed modules. Compared with state-of-the-art methods, our method extracted much more branches while maintaining competitive overall segmentation performance. © Springer Nature Switzerland AG 2020.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Dual-Domain deep prior guided sparse-view CT reconstruction with multi-scale fusion attention

引用

Scientific Reports 2025年第1期15卷 1-20页

作者： Wu, Jia Lin, Jinzhao Jiang, Xiaoming Zheng, Wei Zhong, Lisha Pang, Yu Meng, Hongying Li, Zhangyong School of Communications and Information Engineering Chongqing University of Posts and Telecommunications Chongqing 400065 China School of Medical Information and Engineering Southwest Medical University Luzhou 646000 China Chongqing Engineering Research Center of Medical Electronics and Information Technology Chongqing University of Posts and Telecommunications Chongqing 400065 China Key Laboratory of Intelligent Computing for Advanced Manufacturing Chongqing University of Posts and Telecommunications Chongqing 400065 China School of Optoelectronic Engineering Chongqing University of Posts and Telecommunications Chongqing China Department of Electronic and Electrical Engineering College of Engineering Design and Physical Sciences Brunel University London Uxbridge UB8 3PH United Kingdom

Sparse-view CT reconstruction is a challenging ill-posed inverse problem, where insufficient projection data leads to degraded image quality with increased noise and artifacts. Recent deep learning approaches have shown promising results in CT reconstruction. However, existing methods often neglect projection data constraints and rely heavily on convolutional neural networks, resulting in limited feature extraction capabilities and inadequate adaptability. To address these limitations, we propose a Dual-domain deep Prior-guided Multi-scale fusion Attention (DPMA) model for sparse-view CT reconstruction, aiming to enhance reconstruction accuracy while ensuring data consistency and stability. First, we establish a residual regularization strategy that applies constraints on the difference between the prior image and target image, effectively integrating deep learning-based priors with model-based optimization. Second, we develop a multi-scale fusion attention mechanism that employs parallel pathways to simultaneously model global context, regional dependencies, and local details in a unified framework. Third, we incorporate a physics-informed consistency module based on range-null space decomposition to ensure adherence to projection data constraints. Experimental results demonstrate that DPMA achieves improved reconstruction quality compared to existing approaches, particularly in noise suppression, artifact reduction, and fine detail preservation. © The Author(s) 2025.

关键词： Deep prior Model-based optimization Multi-scale fusion attention Physics-informed consistency Sparse-view CT reconstruction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：