检索结果-内蒙古大学图书馆

ieee/ICME International conference on Complex Medical Engineering, CME

作者： Yujue Wang Department of Chemistry and Chemical Biology University of New Mexico Albuquerque US

Nowadays, the typical tools employed in the diagnosis of the pandemic coronavirus disease, COVID-19, including Real-Time Reverse Transcription Polymerase Chain Reaction (RT-PCR) and Polymerase Chain Reaction (PCR), which are less sensitive, time-consuming, and demanding assistance from expert medical personnel assistance. Computed tomography (CT) as the artificial intelligence (AI) technological utilized in high accurate COVID-19 infection screening in a short amount of time is tremendously helpful. To address those limitations mentioned above, In this paper, a robust, optimized model for detection of the COVID-19 automatically in digital CT images is proposed utilizing the technique based on transfer learning and attention mechanism derived from deep learning. MobileNet-V1 architecture of transfer learning was applied to make the model more lightweight and reduce the computation, while setting to be the pre-trained mode meanwhile. In addition, the attention mechanism of SENet’s called Squeeze-and-Excitation (SE) module was employed to let it learn the significance of various channel features automatically. Two experiments, with transfer learning and attention mechanism technique or not, were employed to assess the function of the model. Noteworthy, the accuracy, precision, recall, and F1-score were 95.97%, 93.47%, 94.27%, and 94.01% respectively. The results reveal that the optimized approach outperform the comparative models.

关键词： COVID-19 Deep learning Computed tomography computational modeling Pulmonary diseases Transfer learning Feature extraction

来源：评论

学校读者我要写书评

暂无评论

ieee Transactions on Pattern Analysis and Machine intelligence Editorial Board

引用

ieee/ACM Transactions on computational biology and bioinformatics 2014年第6期11卷 C2-C2页

Provides a listing of current staff, committee members and society officers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ieee Transactions on Pattern Analysis and Machine intelligence Information for Authors

引用

ieee/ACM Transactions on computational biology and bioinformatics 2014年第6期11卷 C3-C3页

Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

关键词：

来源：评论

学校读者我要写书评

暂无评论

New for 2004

引用

ieee Transactions on Pattern Analysis and Machine intelligence 2004年第2期26卷 288-288页

来源：评论

学校读者我要写书评

暂无评论

Prognostic Value of Brain Tissue Pathological Changes in Patients with Clinically Isolated Syndromes (CIS) Suggestive of Multiple Sclerosis Using Magnetization Transfer Ratio (MTR)

Prognostic Value of Brain Tissue Pathological Changes in Pat...

引用

Annual International conference of the ieee Engineering in Medicine and biology Society (EMBC)

作者： M. Fooladi N. Riyahi Alam M. H. Harirchyan K. Firuznia M. A. Oghabian M. Shakiba B. Rafie M. Bakhtiary Department of Medical Physics & Biomedical Engineering School of Medicine Medical Sciences University of Tehran Tehran Iran Department of Clinical Neurology School of Medicine Medical Sciences University of Tehran Tehran Iran Department of Radiology&Imaging Center School of Medicine Medical Sciences University of Tehran Tehran Iran Research Center for the Science and Technology Imam hospital Medical Sciences University of Tehran Tehran Iran

The aim of this study is to investigate abnormalities in the brain tissue of patients with clinically isolated syndrome (CIS) suggestive of multiple sclerosis (MS). In this method, magnetization transfer ratio (MTR) parameter accompanied with segmentation regional measurements and histogram analysis were used to improve the evaluation of disease progression in CIS patients. Conventional MR imaging protocols such as Tl-weighted, T2- weighted, T2-FLAIR as well as MT-2DSPGR were performed on four CIS patients and four normal subjects. White matter, gray matter and lesion masks were segmented from Tl- weighted images and superimposed on MTR map using FSL software. Lesions were classified into isontense and severely hypointense according to their signal intensities relative to white matter on the Tl-weighted images. MTR parameters of these two lesion types, normal appearing white matter (NAWM) and normal appearing gray matter (NAGM) were analyzed in comparison with those of normal controls. The MTR histograms of NAWM and NAGM were also generated for each segmented brain tissues. A significant reduction was found in mean white matter MTR and the histogram peak position between CIS patients and healthy subjects. The MTR histogram for NAWM showed also a total shift to the left. The MTR value for gray matter in CIS patients was similar to that of controls. Isointense lesions have significantly higher MTR values than severely hypointense lesions. Significant reduction in NAWM-MTR compared to normal subjects shows that pathological changes outside visible lesions on conventional MR images occur among patients with CIS at presentation. Quantitative MT imaging can help to evaluate the extent of global damage in the brain tissue.

关键词： Brain Pathology computational intelligence Society Multiple sclerosis Magnetization Lesions Histograms Image segmentation Magnetic analysis Diseases

来源：评论

学校读者我要写书评

暂无评论

A real-time HFO detection framework for localization of SOZ in iEEG recording of patients with epilepsy

A real-time HFO detection framework for localization of SOZ ...

引用

Annual International conference of the ieee Engineering in Medicine and biology Society (EMBC)

作者： Behrang Fazli Besheli Chandra Prakash Swamy Amir Hossein Ayyoubi Luciano R. F. Branco Gregory A. Worrell Zhiyi Sha Jay R. Gavvala Kai J. Miller Nuri F. Ince Department of Neurologic Surgery Mayo Clinic Rochester MN USA Bioinformatics and Computational Biology Department University of Minnesota Minneapolis MN USA Department of Biomedical Engineering University of Houston Houston TX USA Department of Neurology Mayo Clinic Rochester MN USA Department of Neurology University of Minnesota Minneapolis MN USA Department of Neurology UT Health Houston TX USA

ISBN: (数字)9798350371499

ISBN: (纸本)9798350371505

Interictal High-frequency Oscillation (HFO) between 80-600 Hz in intracranial EEG (iEEG) is a promising biomarker of the epileptogenic zone in individuals with epilepsy. Numerous studies revealed that the resection of channels with a high rate of HFOs correlates with favorable surgical outcomes. Early feedback to clinicians regarding the distribution of HFOs during the iEEG recording, especially after the implantation of electrodes, would be helpful for clinical decisions. However, iEEG recording can easily get corrupted by various factors mimicking real HFOs. This study presents a real-time HFO detection framework within MATLAB/Simulink that exhibits robustness against such pseudo-HFOs. This detector is responsible for identifying the initial pool of candidate HFOs and transmitting them via user datagram protocol (UDP) to an external application. The external application contains a machine learning tool that is utilized for post-processing and isolating the real-HFOs. It is implemented in a graphical user interface (GUI) that provides visual feedback regarding the distribution and waveforms of HFOs. The entire processing pipeline was validated by randomly selecting 10-minute segments of interictal iEEG recordings from 10 subjects. It successfully identifies the seizure onset zone (SOZ) in these subjects, achieving an average accuracy of 65% using the detected Ripples and 74% using the detected events with both Ripples and Fast Ripples. Importantly, the spatio-temporal distribution of detected HFOs in real-time showed more than 98% spatio-temporal similarity index compared to offline analysis. Our framework proves to be an effective tool for the automatic identification of HFOs in real-time with the ability to promptly stream the HFO analysis results and provide early feedback regarding the probable SOZ regions to clinicians for surgical decision-making.

关键词： Visualization Protocols Pipelines Surgery Epilepsy Hafnium oxide Real-time systems Robustness Recording Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

Predicting drug-target interactions using restricted Boltzmann machines

引用

bioinformatics 2013年第13期29卷 126-134页

作者： Wang, Yuhao Zeng, Jianyang Tsinghua Univ Dept Automat Beijing 100084 Peoples R China Tsinghua Univ Inst Interdisciplinary Informat Sci Beijing 100084 Peoples R China

Motivation: In silico prediction of drug-target interactions plays an important role toward identifying and developing new uses of existing or abandoned drugs. Network-based approaches have recently become a popular tool for discovering new drug-target interactions (DTIs). Unfortunately, most of these network-based approaches can only predict binary interactions between drugs and targets, and information about different types of interactions has not been well exploited for DTI prediction in previous studies. On the other hand, incorporating additional information about drug-target relationships or drug modes of action can improve prediction of DTIs. Furthermore, the predicted types of DTIs can broaden our understanding about the molecular basis of drug action. Results: We propose a first machine learning approach to integrate multiple types of DTIs and predict unknown drug-target relationships or drug modes of action. We cast the new DTI prediction problem into a two-layer graphical model, called restricted Boltzmann machine, and apply a practical learning algorithm to train our model and make predictions. Tests on two public databases show that our restricted Boltzmann machine model can effectively capture the latent features of a DTI network and achieve excellent performance on predicting different types of DTIs, with the area under precision-recall curve up to 89.6. In addition, we demonstrate that integrating multiple types of DTIs can significantly outperform other predictions either by simply mixing multiple types of interactions without distinction or using only a single interaction type. Further tests show that our approach can infer a high fraction of novel DTIs that has been validated by known experiments in the literature or other databases. These results indicate that our approach can have highly practical relevance to DTI prediction and drug repositioning, and hence advance the drug discovery process.

关键词： Algorithms Artificial intelligence Computer Simulation Drug Discovery/methods Drug Repositioning Humans Pharmaceutical Preparations/metabolism Proteins/drug effects Proteins/metabolism

来源：评论

学校读者我要写书评

暂无评论

Growing Bayesian network models of gene networks from seed genes

引用

bioinformatics 2005年第Sup2期21卷 224-229页

作者： Peña, JM Björkegren, J Tegnér, J Linkoping Univ Dept Phys & Measurement Technol S-58183 Linkoping Sweden Karolinska Inst Ctr Genom & Bioinformat S-17177 Stockholm Sweden

Motivation: For the last few years, Bayesian networks (BNs) have received increasing attention from the computational biology community as models of gene networks, though learning them from gene-expression data is problematic. Most gene-expression databases contain measurements for thousands of genes, but the existing algorithms for learning BNs from data do not scale to such high-dimensional databases. This means that the user has to decide in advance which genes are included in the learning process, typically no more than a few hundreds, and which genes are excluded from it. This is not a trivial decision. We propose an alternative approach to overcome this problem. Results: We propose a new algorithm for learning BN models of gene networks from gene-expression data. Our algorithm receives a seed gene S and a positive integer R from the user, and returns a BN for the genes that depend on S such that less than R other genes mediate the dependency. Our algorithm grows the BN, which initially only contains S, by repeating the following step R + 1 times and, then, pruning some genes;find the parents and children of all the genes in the BN and add them to it. Intuitively, our algorithm provides the user with a window of radius R around S to look at the BN model of a gene network without having to exclude any gene in advance. We prove that our algorithm is correct under the faithfulness assumption. We evaluate our algorithm on simulated and biological data (Rosetta compendium) with satisfactory results.

关键词： Algorithms Artificial intelligence Bayes Theorem Computer Simulation Gene Expression/physiology Gene Expression Profiling/methods Models, Genetic Neural Networks (Computer) Pattern Recognition, Automated/methods Proteome/metabolism Signal Transduction/genetics Signal Transduction/physiology

来源：评论

学校读者我要写书评

暂无评论

A predictor for toxin-like proteins exposes cell modulator candidates within viral genomes

引用

bioinformatics 2010年第18期26卷 i482-i488页

作者： Naamati, Guy Askenazi, Manor Linial, Michal Hebrew Univ Jerusalem Dept Biol Chem Sudarsky Ctr Computat Biol IL-91905 Jerusalem Israel Hebrew Univ Jerusalem Sch Comp Sci & Engn IL-91905 Jerusalem Israel Dana Farber Canc Inst Blais Prote Ctr Boston MA 02115 USA

Motivation: Animal toxins operate by binding to receptors and ion channels. These proteins are short and vary in sequence, structure and function. Sporadic discoveries have also revealed endogenous toxin-like proteins in non-venomous organisms. Viral proteins are the largest group of quickly evolving proteomes. We tested the hypothesis that toxin-like proteins exist in viruses and that they act to modulate functions of their hosts. Results: We updated and improved a classifier for compact proteins resembling short animal toxins that is based on a machine-learning method. We applied it in a large-scale setting to identify toxin-like proteins among short viral proteins. Among the similar to 26 000 representatives of such short proteins, 510 sequences were positively identified. We focused on the 19 highest scoring proteins. Among them, we identified conotoxin-like proteins, growth factors receptor-like proteins and anti-bacterial peptides. Our predictor was shown to enhance annotation inference for many 'uncharacterized' proteins. We conclude that our protocol can expose toxin-like proteins in unexplored niches including metagenomics data and enhance the systematic discovery of novel cell modulators for drug development.

关键词： Algorithms Amino Acid Sequence Animals Artificial intelligence Base Sequence Conotoxins/chemistry Metagenomics Molecular Sequence Data Protein Folding Protein Structure, Tertiary Proteome Toxins, Biological/physiology Toxins, Biological/toxicity Viral Proteins/chemistry Viral Proteins/physiology Viral Proteins/toxicity

来源：评论

学校读者我要写书评

暂无评论

TAME-Faster R-CNN model for Image-based Tea Diseases Detection

TAME-Faster R-CNN model for Image-based Tea Diseases Detecti...

引用

Canadian conference on Electrical and Computer Engineering (CCECE)

作者： Tiancheng Liu Ling Bai Rakiba Rayhana Xiuguo Zou Zheng Liu School of Engineering University of British Columbia BC Kelowna Canada College of Artificial Intelligence Nanjing Agricultural University Nanjing China

ISBN: (数字)9798350371628

ISBN: (纸本)9798350371635

Tea, one of the most consumed non-alcoholic beverages in the world, plays an essential role in the agricultural economy. Nevertheless, it is threatened by various diseases, resulting in critical yields and economic losses. Nowadays, image processing techniques and machine vision algorithms are used to detect tea diseases. However, the existing techniques do not consider the varying lighting (such as shadow) conditions in the data set, which makes the techniques less efficient and robust. Therefore, this paper aims to address these issues by proposing an efficient technique called, TAME-Faster R-CNN. The proposed method combines a trainable Attention Mechanism for Explanations (TAME) module with the backbone network of the Faster R-CNN framework to detect three types of tea diseases (Anthracnose, Brown leaf spot, and Tea white scab). The experimental results and analysis show that the proposed algorithm achieved mAP values of 98.3% to detect Anthracnose, 64.8% to detect Brown leaf spots, and 75.6% to detect white scab diseases and performed better than the state-of-the-art technique. Total mAP is almost 10%, 23% and 1% higher compared to Yolov5, Yolov7 and original Faster R-CNN, respectively. Compared to the original framework of Faster R-CNN, TAME-Faster R-CNN with ResNet101 has improved the precision and F1 score on average by 6% and 4.3%, respectively. Hence, the integration of this technique can effectively detect tea lesion diseases.

关键词： YOLO Economics Attention mechanisms computational modeling Plants (biology) Machine vision Lighting

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：