检索结果-内蒙古大学图书馆

2022生物物理大会

作者： Liqing Liu Shuxin Yang Yang Liu Xixia Li Junjie Hu Li Xiao Tao Xu National Laboratory of Biomacromolecules Institute of Biophysics Chinese Academy of Sciences Center for Biological Imaging Institute of Biophysics Chinese Academy of Sciences Key Laboratory of Intelligent Information Processing Institute of Computing TechnologyChinese Academy of Sciences School of Computer and Control Engineering University of Chinese Academy of Sciences College of Life Science University of Chinese Academy of Sciences Ningbo HuaMei Hospital University of Chinese Academy of Sciences School of Biomedical Engineering Guangzhou Medical University

Membrane contact site (MCS)-mediated organelle interactions plays essential roles in the cell. Quantitative analysis of the MCS reveals vital clues for cellular responses under various physiological and pathological conditions. However, an efficient tool is lacking. Here, we developed "DeepContact", a deep learning protocol for optimizing organelle segmentation and contact analysis based on label-free electron microscopy(EM).

关键词： DeepContact High throughput quantification of membrane contact site based on electron microscopy imaging

来源：评论

学校读者我要写书评

暂无评论

I/O efficient core maintenance in large graphs

引用

Expert Systems with Applications 2025年 288卷

作者： Jie Dai Yixiao Liu Yuan Li Fazal Wahab Yuhai Zhao School of Computer Science and Engineering Northeastern University Shenyang 110819 China Key Laboratory of Intelligent Computing of Medical Images Ministry of Education Northeastern University Shenyang 110819 China School of Information Science and Technology North China University of Technology Beijing 100144 China

Coreness is a fundamental concept in graph studies and has board applications in areas such as node importance analysis. Core maintenance aims to efficiently updating the coreness when the graph is changed. The existing multi-edge update strategies follow the single-edge update framework and lack the systematic description of the coreness change under the multi-edge update, leading to high processing complexity. Meanwhile, these approaches are generally assume that the entire graph resides in memory, which makes it difficult to deal with the large graphs. To address these limitations, we investigate at both theoretical and algorithmic aspects. Theoretically, we establish the complete theory of core maintenance, proving the affected nodes and change rule of coreness under the multi-edge update. Algorithmically, we design I/O efficient core maintenance algorithms based on the semi-external model, which adopts the convergence paradigm that iteratively refines node coreness from initial upper limit to the exact value. Specifically, for the multi-edge insertion, we determine the affected nodes and their upper limit of coreness based on the proposed theory, and then utilize the convergence paradigm to realize the exact calculation of coreness. For the multi-edge deletion, we efficiently extend the existing single-edge method. In addition, for the complex multi-edge insertion scenario, we further propose three optimization strategies to reduce the search space, search time, and coreness convergence time. Extensive experiments conducted on 7 real datasets show that our proposed algorithms outperform the existing works. For instance, when inserting 10,000 edges into the SW dataset, the base algorithm is 436 % faster in terms of running time and reduces I/O consumption by 97.9 %. The best-optimized algorithm is 2,764 % faster and reduces I/O consumption by 98.8 %. Furthermore, the memory consumption is only 297 MB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Scalar coupling constant prediction using graph embedding local attention encoder

arXiv

引用

arXiv 2020年

作者： Jian, Caiqing Cheng, Xinyu Zhang, Jian Wang, Lihui Key Laboratory of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province College of Computer Science and Technology Guizhou University Guiyang550025 China

Scalar coupling constant (SCC) plays a key role in the analysis of three-dimensional structure of organic matter, however, the traditional SCC prediction using quantum mechanical calculations is very time-consuming. To calculate SCC efficiently and accurately, we proposed a graph embedding local self-attention encoder (GELAE) model, in which, a novel invariant structure representation of the coupling system in terms of bond length, bond angle and dihedral angle was presented firstly, and then a local self-attention module embedded with the adjacent matrix of a graph was designed to extract effectively the features of coupling systems, finally, with a modified classification loss function, the SCC was predicted. To validate the superiority of the proposed method, we conducted a series of comparison experiments using different structure representations, different attention modules, and different losses. The experimental results demonstrate that, compared to the traditional chemical bond structure representations, the rotation and translation invariant structure representations proposed in this work can improve the SCC prediction accuracy;with the graph embedded local self-attention, the mean absolute error (MAE) of the prediction model in the validation set decreases from 0.1603 Hz to 0.1067 Hz;using the classification based loss function instead of the scaled regression loss, the MAE of the predicted SCC can be decreased to 0.0963 HZ, which is close to the quantum chemistry standard on CHAMPS dataset. Copyright © 2020, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Pulmonary vascular segmentation algorithm based on fractional differential enhancement

Pulmonary vascular segmentation algorithm based on fractiona...

引用

International Conference on Mechatronics and intelligent Robotics, ICMIR 2018

作者： Xiao, Yuqing Tan, Wenjun Zhou, Qing Ji, Yu College of Computer Science and Engineering Northeastern University Shenyang China Key Laboratory of Medical Image Computing of Ministry of Education Northeastern University Shenyang China

ISBN: (纸本)9783030002138

With one of the highest incidence rates, Lung cancer accounts for 26.8% of cancer deaths. Due to the pulmonary segment having a separate pulmonary artery and pulmonary vein, the division of the pulmonary artery and pulmonary vein is critical for path planning in surgery of segmental resection of the lung. As a result of this previous research, CT images provided are used as data sources. Combined with the anatomical structure characteristics of pulmonary vessels, the following segmentation methods of the pulmonary artery and pulmonary vein are studied in this thesis. (a) image segmentation and center line extraction is introduced to lung parenchyma segmentation, based on the 3D growth region method. (b) Morphology used to fill the holes in the segmented pulmonary parenchyma, operations made between filled images and the original image to obtain the lungs filled with pulmonary vessels, fractional differential operator designed to enhance the gray level of the small blood vessels, and complete pulmonary blood vessel segmented by local optimal threshold. (c) Algorithm for extracting the centerline of lung blood vessels based on two distance field is researched and arteriovenous conglutination part of the pulmonary centerline detected and removed automatically to obtain the vascular subtree. Finally, Using topological anatomy of the pulmonary artery and vein, the vascular tree classification algorithm based on the subtree leaf node matching is studied along with the pulmonary blood vessels which are grown by region growth method through the center line. The pulmonary artery and vein are then extracted from the chest CT image. The experimental results show that this proposed pulmonary artery and vein segmentation method has high accuracy and short operation time, and has good clinical application value. This newly developed algorithm and implementation will have important applications for treatment planning, dose calculations and treatment validation during cancer radiat

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

Knowledge Distillation Based on Adaptive Learning and Channel Amplification Features for PolSAR image Classification

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Ronghua Shang Mingwei Hu Wenzheng Li Lei Liu Weitong Zhang Jie Feng Songhua Xu Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education School of Artificial Intelligence Xidian University Xi’an China Guangzhou Institute of Technology Xidian University Guangzhou China Key Laboratory of Electronic Information Countermeasure and Simulation Technology Ministry of Education Xidian University Xi’an China Institute of Medical Artificial Intelligence The Second Affiliated Hospital of Xi’an Jiaotong University Xi’an China

The models currently used for polarimetric synthetic aperture radar (PolSAR) image classification tasks have problems, such as complex network structures, poor distinction of detailed features, and fixed loss weights during the training process. In response to these problems, this article proposes a PolSAR image classification method based on knowledge distillation using adaptive learning and channel amplification features. First, this article builds a knowledge distillation framework for PolSAR. Using a teacher network trained in advance that can acquire global knowledge to guide the student. This framework reduces the computational complexity and improves the classification accuracy of the student. Then, an adaptive loss weight learning mechanism is designed, which sets the weight of the Kullback-Leibler (KL) divergence loss during training into a learnable mode. The weight can be automatically adjusted according to the actual training situation of the student. Finally, a scheme for channel amplification to enhance features is proposed. This scheme obtains channel weights based on the student’s feature map information. These weights are amplified, strengthening the network’s ability to obtain feature information. Compared with the five PolSAR image classification algorithms, the method proposed in this article uses lower computational complexity to obtain higher classification accuracy on the Flevoland, San Francisco, and Xi’an datasets.

关键词： Training image classification Feature extraction Adaptation models Computational modeling Computational complexity Scattering Data mining Accuracy Knowledge engineering

来源：评论

学校读者我要写书评

暂无评论

Pedestrian partial occlusion detection method based on improved YOLO network structure

Pedestrian partial occlusion detection method based on impro...

引用

2019 International Conference on intelligent computing, Automation and Systems, ICICAS 2019

作者： Chen, Guangxi Zhan, Baoying Tang, Min Xie, Chen Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin China School of Mathematics and Computing Science Guilin University of Electronic Technology Guilin China

ISBN: (纸本)9781728161068

The problem of pedestrian detection accuracy is low due to partial occlusion of pedestrians in practical application scenarios, a pedestrian occlusion detection method based on improved YOLOv3 network structure is proposed. By improving the activation function and clustering method to strengthen the feature output layer of the shallow layer of the network, which effectively make the accuracy of detection better. The experimental results show that the mAP of proposed method is improved by 2.63 percent in the INRIA data set, and the improved network has good robustness and generalization ability, and guarantees the real-time of detection. © 2019 IEEE.

关键词： Chemical activation

来源：评论

学校读者我要写书评

暂无评论

Arbitrary shape text detection method based on improved PSE network structure

Arbitrary shape text detection method based on improved PSE ...

引用

IEEE International Conference of Safety Produce Informatization (IICSPI)

作者： Chen Xie Guangxi Chen Min Tang Qiaochu Li Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin China School of Mathematics and Computing Science Guilin University of Electronic Technology Guilin China

ISBN: (数字)9781728177380

ISBN: (纸本)9781728177397

The problem of low efficiency of text detection in natural scenes due to the Curved text or the multi-directional text, a arbitrary shape text detection method based on PSE network structure is proposed. By changing the backbone network of the original PSE and adopting a new lightweight network as backbone, and the post-processing method of the original network is optimized, which effectively improves the speed without losing the original accuracy. The experimental results show that the speed of the proposed method is 5.81 FPS/s compared with the original PSE on CTW1500 data set. The improved network has good robustness and generalization ability, and has better overall performance.

关键词： Shape Conferences Production Robustness

来源：评论

学校读者我要写书评

暂无评论

An Artificial Intelligence-Based Ecological Index for Prognostic Evaluation of Colorectal Cancer

SSRN

引用

SSRN 2022年

作者： Chen, Qicong Cai, Ming Fan, Xinjuan Liu, Wenbin Fang, Gang Yao, Su Xu, Yao Li, Qian Zhao, Yingnan Zhao, Ke Liu, Zaiyi Chen, Zhihua Institute of Computing Science and Technology Guangzhou University Guangzhou China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangzhou China Department of Radiology Guangdong Provincial People's Hospital Guangdong Academy of Medical Sciences Guangzhou China Department of Pathology The Sixth Affiliated Hospital of Sun Yat-sen University Guangzhou China Department of Pathology Guangdong Provincial People's Hospital Guangdong Academy of Medical Sciences Guangzhou China Guangdong Cardiovascular Institute Guangdong Provincial People's Hospital Guangdong Academy of Medical Sciences Guangzhou China

Background and Objective: In the tumor microenvironment (TME), the active interaction between tumor cells and immune cells is crucial for predicting the prognosis of colorectal cancer. This study proposed a Morisita-Horn ecological index to quantify the spatial associations between tumor cells and immune cells in the TME by artificial intelligence (AI)-based method from immunohistochemistry (IHC)-stained whole-slide images (WSIs) of colorectal cancer (CRC) *** and methods: We used deep learning technology and traditional computer segmentation methods to segment the tumor nuclei, immune nuclei, and stroma nuclei in tumor regions in IHC-stained WSIs. The Morisita-Horn ecological index (Mor-index) was used to evaluate the spatial associations between the tumor cells and immune cells in the TME of CRC patients. A discovery cohort (N = 432) and validation cohort (N = 137) were used to evaluate the prognostic value of Mor-index for overall survival (OS).Results: Experiments were conducted on two data sets of 569 patients to demonstrate the efficacy of the method. Compared to other studies, our method is not only superior to the QuPath tool but also produces better segmentation results with an accuracy of 0.85. Mor-index was quantified automatically by our method. Survival analysis indicated that the higher Mor-index correlated with better OS in the discovery cohorts (HR for high vs. low 0.49, 95% CI 0.27–0.77, P = 0.0014) and validation cohort (0.21, 0.10–0.46, Conclusion: This study provided a novel AI-based approach to segmenting various nuclei in the TME. The Mor-index can reflect the immune status of CRC patients and is associated with favorable survival. Thus, Mor-index can potentially make a significant role in aiding clinical prognosis and decision-making. © 2022, The Authors. All rights reserved.

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

Design and Evaluation of an Exergame System of Knee with the Azure Kinect

Design and Evaluation of an Exergame System of Knee with the...

引用

2021国际计算机前沿大会

作者： Guangjun Wang Ming Cheng Xueshu Wang Yi Fan Xin Chen Liangliang Yao Hanyuan Zhang Zuchang Ma Hefei Institutes of Physical Science Chinese Academy of Sciences University of Science and Technology of China The University Key Laboratory of Intelligent Perception and Computing of Anhui Province Anqing Normal University Department of Sports Medicine and Arthroscopic Surgery The First Affiliated Hospital of Anhui Medical University

Timely and effective knee function evaluation and knee exercises promote the prevention and self-management of knee diseases. In this paper, a Kinectbased exergame system is proposed to assess and train the knee function. Azure Kinect was used to capture and generate 3 D models of the user and immerse them in an interactive virtual environment. The software included three functional modules: knee function evaluation, Knee exercises game, and Comprehensive evaluation. The stand, step, leg lift, and squat were selected for knee function evaluation and exercises. Twenty volunteers participated in the experiment. Intraclass correlation coefficients(ICCs) were calculated to assess the reliability of kinematic measurements of knee angles during the movements. The ICC of these movements were stand(ICC = 0.987), step(ICC = 0.997), left leg lift(ICC =0.981), right leg lift(ICC = 0.990), stand(ICC = 0.998). The results show that the test-retest reliability is high. It means that the motion capture data is effective and the data obtained by Kinect is stable.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Automatic Classification of Arrhythmias by Residual Network and BiGRU With Attention Mechanism

Automatic Classification of Arrhythmias by Residual Network ...

引用

Computers in Cardiology (CinC)

作者： Runnan He Kuanquan Wang Na Zhao Qiang Sun Yacong Li Qince Li Henggui Zhang School of Computer Science and Technology Harbin Institute of Technology (HIT) Harbin Heilongjiang China Beijing Electric Power Hospital Beijing China International Laboratory for Smart Systems and Key Laboratory of Intelligent of Computing in Medical Image Ministry of Education Northeastern University Shenyang China

ISBN: (数字)9781728173825

ISBN: (纸本)9781728111056

Aims: Over the last decade, many attempts have been implemented for automatic detection of cardiac arrhythmias, however, their performances are still not ideal due to unreliable extracted features of designed models or limited small public datasets. In this study, we investigate automatic detection of arrhythmias from 12-lead electrocardiogram (ECG) recordings using an attention-based Res-BiGRU model. Methods: We train a deep neural networks (DNNs) to identify eight types of arrhythmias. The constructed model contains residual convolutional modules and bidirectional Gated Recurrent Unit (BiGRU) layers to extract features from preprocessed ECG signals. The mechanism of attention is applied to learn an attention distribution on the list of extracted features, and sum the weighted features by the attention into a single feature vector, that is responsible for the final classification. We train our model on the training set of the PhysioNet/computing in Cardiology Challenge 2020 with time length ranging from 6 s to 60 s. The proposed model can handle signals with variable lengths. Results: The overall score with five-cross validation of training set is 0.543 for our team (DeepHeart), which demonstrates a good efficacy of our model, demonstrating that it may have potential practical applications.

关键词： Training Convolution Computational modeling Electrocardiography Feature extraction Cardiology Residual neural networks

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：