检索结果-内蒙古大学图书馆

MatConvNet-Based Fast Method for Cervical MR Images Classification

International Conference on Computational Intelligence: Theories, Applications and Future Directions, ICCI 2017

作者： Verma, Gyanendra K. Lather, Jagdeep S. Kaushal, Ankur Department of Computer Engineering National Institute of Technology Kurukshetra Kurukshetra136119 India Department of Electrical Engineering National Institute of Technology Kurukshetra Kurukshetra136119 India School of Biomedical Engineering National Institute of Technology Kurukshetra Kurukshetra136119 India

ISBN: (纸本)9789811311345

The deep convolutional neural network (CNN) has been successfully used to obtain high-level representation in various applications of computer vision problems. However, in the field of medical imaging there are not sufficient images available to train a deep CNN. Therefore, we have used a pre-trained deep CNN model for classification of cervical cancer MR images. In this paper, we have proposed MatConvNet-based CNN model to extract features from pre-trained CNN for classification. The vgg-f architecture is deployed to extract the image features. We have evaluated our proposed system with benchmark cervical cancer database obtained from Tumor Cancer Imaging Archive (TCIA). We got the promising result with 98.9% accuracy that is beyond the methods reported in the literature. © 2019, Springer Nature Singapore Pte Ltd.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A systematic review of EEG source localization techniques and their applications on diagnosis of brain abnormalities

arXiv

引用

arXiv 2019年

作者： Asadzadeh, Shiva Rezaii, Tohid Yousefi Beheshti, Soosan Delpak, Azra Meshgini, Saeed Department of Biomedical Engineering Faculty of Electrical and Computer Engineering University of Tabriz Department of Biomedical Engineering Faculty of Electrical and Computer Engineering University of Tabriz Tabriz Iran Department of Electrical and Computer Engineering Ryerson University Tabriz University of Medical Sciences

In recent years, multiple noninvasive imaging modalities have been used to develop a better understanding of the human brain functionality, including positron emission tomography, single-photon emission computed tomography, and functional magnetic resonance imaging, all of which provide brain images with millimeter spatial resolutions. Despite good spatial resolution, time resolution of these methods are poor and values are about seconds. Electroencephalography (EEG) is a popular non-invasive electrophysiological technique of relatively very high time resolution which is used to measure electric potential of brain neural activity. Scalp EEG recordings can be used to perform the inverse problem in order to specify the location of the dominant sources of the brain activity. In this paper, EEG source localization research is clustered as follows: solving the inverse problem by statistical method (37.5%), diagnosis of brain abnormalities using common EEG source localization methods (18.33%), improving EEG source localization methods by non-statistical strategies (3.33%), investigating the effect of the head model on EEG source imaging results (12.5%), detection of epileptic seizures by brain activity localization based on EEG signals (20%), diagnosis and treatment of ADHD abnormalities (8.33%). Among the available methods, minimum norm solution has shown to be very promising for sources with different depths. This review investigates diseases that are diagnosed using EEG source localization techniques. In this review we provide enough evidence that the effects of psychiatric drugs on the activity of brain sources have not been enough investigated, which provides motivation for consideration in the future research using EEG source localization methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Machine learning based potentiating impacts of 12-lead ECG for classifying paroxysmal versus non-paroxysmal atrial fibrillation

引用

International Journal of Arrhythmia 2022年第1期23卷 1-9页

作者： Sungsoo Kim Alan C. Bovik Sohee Kwon Sung-Hwi Hong JunYong Kim Boyoung Joung Hui-Nam Pak Moon-Hyeong Lee Mia K. Markey Hye Jin Hwang Junbeom Park Electrical and Computer Engineering The University of Texas at Austin Austin USA Deparment of Internal Medicine College of Medicine Yonsei University Seoul South Korea Department of Epidemiology Harvard T. H. Chan School of Public Health Boston MA USA Harvard University Cambridge USA Department of Biomedical Engineering The University of Texas at Austin Austin USA Department of Imaging Physics The University of Texas MD Anderson Cancer Center Houston USA Cardiovascular Division Department of Medicine Beth Israel Deaconess Medical Center Boston MA USA Department of Cardiology College of Medicine Ewha Womans University 1071 Annyangcheon-ro Yangcheon-gu Seoul Republic of Korea

BackgroundConventional modality requires several days observation by Holter monitor to differentiate atrial fibrillation (AF) between Paroxysmal atrial fibrillation (PAF) and Non-paroxysmal atrial fibrillation (Non-PA... 详细信息

Background

Conventional modality requires several days observation by Holter monitor to differentiate atrial fibrillation (AF) between Paroxysmal atrial fibrillation (PAF) and Non-paroxysmal atrial fibrillation (Non-PAF). Rapid and practical differentiating approach is needed.

Objective

To develop a machine learning model that observes 10-s of standard 12-lead electrocardiograph (ECG) for real-time classification of AF between PAF versus Non-PAF.

Methods

In this multicenter, retrospective cohort study, the model training and cross-validation was performed on a dataset consisting of 741 patients enrolled from Severance Hospital, South Korea. For cross-institutional validation, the trained model was applied to an independent data set of 600 patients enrolled from Ewha University Hospital, South Korea. Lasso regression was applied to develop the model.

Results

In the primary analysis, the Area Under the Receiver Operating Characteristic Curve (AUC) on the test set for the model that predicted AF subtype only using ECG was 0.72 (95% CI 0.65–0.80). In the secondary analysis, AUC only using baseline characteristics was 0.53 (95% CI 0.45–0.61), while the model that employed both baseline characteristics and ECG parameters was 0.72 (95% CI 0.65–0.80). Moreover, the model that incorporated baseline characteristics, ECG, and Echocardiographic parameters achieved an AUC of 0.76 (95% CI 0.678–0.855) on the test set.

Conclusions

Our machine learning model using ECG has potential for automatic differentiation of AF between PAF versus Non-PAF achieving high accuracy. The inclusion of Echocar

关键词： Machine learning 12 leads surface electrocardiogram Atrial fibrillation

来源：评论

学校读者我要写书评

暂无评论

Dispensed transformer network for unsupervised domain adaptation

arXiv

引用

arXiv 2021年

作者： Li, Yunxiang Li, Jingxiong Dan, Ruilong Wang, Shuai Jin, Kai Zeng, Guodong Wang, Jun Pan, Xiangji Zhang, Qianni Zhou, Huiyu Jin, Qun Wang, Li Wang, Yaqi College of Computer Science and Technology Hangzhou Dianzi University Hangzhou China Artificial Intelligence and Biomedical Image Analysis Lab Westlake University Hangzhou China School of Mechanical Electrical and Information Engineering Shandong University Weihai China Department of Ophthalmology Hospital of Zhejiang University Hangzhou China sitem Center for Translational Medicine and Biomedical Entrepreneurship University of Bern Bern Switzerland School of Biomedical Engineering Shanghai Jiao Tong University Shanghai China School of Electronic Engineering and Computer Science Queen Mary University of London London United Kingdom School of Computing and Mathematical Sciences University of Leicester United Kingdom Department of Human Informatics and Cognitive Sciences Faculty of Human Sciences Waseda University Tokyo Japan Department of Radiology and Biomedical Research Imaging Center University of North Carolina at Chapel Hill Chapel Hill United States College of Media Engineering Communication University of Zhejiang Hangzhou China

Accurate segmentation is a crucial step in medical image analysis and applying supervised machine learning to segment the organs or lesions has been substantiated effective. However, it is costly to perform data annotation that provides ground truth labels for training the supervised algorithms, and the high variance of data that comes from different domains tends to severely degrade system performance over cross-site or cross-modality datasets. To mitigate this problem, a novel unsupervised domain adaptation (UDA) method named dispensed Transformer network (DTNet) is introduced in this paper. Our novel DTNet contains three modules. First, a dispensed residual transformer block is designed, which realizes global attention by dispensed interleaving operation and deals with the excessive computational cost and GPU memory usage of the Transformer. Second, a multi-scale consistency regularization is proposed to alleviate the loss of details in the low-resolution output for better feature alignment. Finally, a feature ranking discriminator is introduced to automatically assign different weights to domain-gap features to lessen the feature distribution distance, reducing the performance shift of two domains. The proposed method is evaluated on large fluorescein angiography (FA) retinal nonperfusion (RNP) cross-site dataset with 676 images and a wide used cross-modality dataset from the MM-WHS challenge. Extensive results demonstrate that our proposed network achieves the best performance in comparison with several state-of-the-art techniques. © 2021, CC BY.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Fast and cross-vendor OpenCL-based implementation for voxelization of triangular mesh models

引用

computer-Aided Design and Applications 2018年第6期15卷 852-862页

作者： Faieghi, Mohammadreza Tutunea-Fatan, O. Remus Eagleson, Roy Biomedical Engineering Western University Canada Mechanical and Materials Engineering Western University Canada Electrical and Computer Engineering Western University Canada

As the open standard for parallel programming of heterogeneous systems, OpenCL has been used in this study in the context of a particular and intensive computing task, namely the voxelization of tessellated objects. For this purpose, OpenCL platform has been utilized to develop a parallelized voxelization algorithm that relies on a fast and efficient triangular mesh facet/cube overlapping test. The extensive numerical tests conducted with heterogeneous hardware configurations on geometric objects of varying complexities, mesh/domain sizes, and voxel resolutions suggest that up to 99.6% or 260 times decrease in the computation time can be obtained when GPU- or CPU-based parallelized techniques are used instead of the conventional single-thread CPU approach. Future developments will attempt the integration of the current implementation into a virtual orthopaedic surgery platform. © 2018, © 2018 CAD Solutions, LLC.

关键词： Parallel programming

来源：评论

学校读者我要写书评

暂无评论

Advanced flexible brain-computer interfaces and devices for the exploration of neural dynamics

Brain‐X

引用

Brain‐X 2024年第4期2卷

作者： Pancheng Zhu Mengxia Yu Mingzheng Wu Yiyuan Yang N.1 Institute for Health National University of Singapore Singapore Singapore Department of Electrical and Computer Engineering National University of Singapore Singapore Singapore Querrey Simpson Institute for Bioelectronics Northwestern University Evanston Illinois USA Department of Biomedical Engineering National University of Singapore Singapore Singapore Institute for Digital Medicine National University of Singapore Singapore Singapore

The rapid advancement of flexible neural interfaces and devices is revolutionizing our ability to explore the neural foundations of consciousness, intelligence, and behavior. Cutting-edge developments in materials science and system-level integration are significantly enhancing the spatiotemporal resolution of neural signal acquisition and modulation, paving the way for next-generation brain-computer interfaces. These technologies enable unprecedented investigations into the causal relationships between neural dynamics and behaviors in freely moving subjects, offering new insights into various neurocognitive domains. The integration of artificial intelligence and brain organoids with neuroscience research promises to further decode complex neural signals, deepening our understanding of multilevel neural dynamics. Beyond their scientific implications, these innovations also offer transformative possibilities for the diagnosis, treatment, and management of neurological and psychiatric disorders. This perspective paper examines how flexible neural interfaces overcome the limitations of traditional neurotechnology, their potential impact on neural research, and their promising applications in treating neurological and psychiatric disorders, while also considering the ethical implications and future challenges in this rapidly evolving field.

关键词： brain-computer interface brain organoid flexible neural interface neural dynamics neurological disease

来源：评论

学校读者我要写书评

暂无评论

A New 3D Imaging Technique Integrating Ultrafast Compounding, Hadamard Encoding, and Reconfigurable Fresnel Lensing, demonstrated on a 128-Element, Crossed Electrode Endoscope

A New 3D Imaging Technique Integrating Ultrafast Compounding...

引用

IEEE International Ultrasonics Symposium

作者： Katherine Latham Christopher Samson Jeffrey Woodacre Eric Simpson Roger Zemp Jeremy A. Brown Biomedical Engineering Dalhousie University Halifax Canada Electrical and Computer Engineering University of Alberta Edmonton Canada Biomedical Engineering Electrical and Computer Engineering Dalhousie University Halifax NS Canada

Crossed electrode arrays address some the challenges associated with 3D ultrasound imaging because of the significant reduction in the number of elements (2N vs. N~2). However, creating a two-way focused 3D image in real-time is difficult with these arrays because azimuth and elevation dimensions cannot be beamformed at the same time. We have developed a new 3D imaging approach that uses the flexibility of bias sensitive substrates to create a high-quality elevation focus on a crossed electrode array. The principle behind this technique is to perform conventional compound imaging with an azimuth set of electrodes, while implementing a bias controllable elevation lens with an elevation set of electrodes. On transmit, the biases are chosen to mimic a Fresnel lens. Then, on receive, Hadamard coding is implemented along the elevation dimension. After decoding, we gain the RF data for each element across the elevation aperture even though there is effectively only one channel in that dimension. A 30MHz, 64×64 element crossed electrode relaxor array was fabricated on a semi-kerfed, electrostrictive substrate and was used to demonstrate the performance of the imaging technique. The -6dB beamwidths were simulated to be 155μm and 170μm in the azimuth and elevation direction respectively and the secondary lobe levels were suppressed below -50dB. Images were generated of a wire phantom to confirm the performance of the elevational focus with good agreement between simulation and experiment.

关键词： Imaging Electrodes Three-dimensional displays Azimuth Lenses Apertures Error correction

来源：评论

学校读者我要写书评

暂无评论

Investigating the use of uni-directional and bi-directional long short-term memory models for automatic sleep stage scoring

引用

Informatics in Medicine Unlocked 2020年 20卷

作者： Fraiwan, Luay Alkhodari, Mohanad Department of Electrical and Computer Engineering Abu Dhabi University Abu Dhabi United Arab Emirates Department of Biomedical Engineering Jordan University of Science and Technology Irbid Jordan

In this paper, a study is conducted to investigate the use of a Long Short-Term Memory (LSTM) learning system in automatic sleep stage scoring. The developed algorithm will automatically learn to classify sleep stages from any acquired sleep signals data-set. This allows to resolve the difficulties that are facing experts in manual sleep stage scoring. A total of 39 Polysomnogram (PSG) recordings acquired from the online PhysioNet Sleep-EDF database are used in this study. The PSG recordings are chosen to be from the EEG Fpz-Cz signals only. The database comes with annotation files that include expert manual stage scoring based on the Rechtschaffen & Kales (R&K) scoring manual. The obtained signals go initially through a pre-processing procedure where sleep stages signals are extracted, normalized, and filtered. The resulting sleep signals are trained using a k-fold cross-validation scheme of 10-folds. Prior to the training and classification process, the LSTM network architecture is built using Uni- and Bi-directional structures to utilize both the forward and backward chains of data sequences. At the end, the developed algorithm performance is evaluated and a complete performance summary table is provided relative to other State-of-the-Art deep learning studies. The performance of this study is evaluated initially without the merging of S3 and S4 sleep stages following the R&K manual, which is considered challenging due to the minor differences between the signals. Then, the performance is evaluated following the recent American Academy of Sleep Medicine (AASM) scoring manual with the merging of the two stages as N3. The developed algorithm achieved higher results using the Bi-directional LSTM. In addition, it achieved the highest accuracy among all other studies in the field with 97.28%. Furthermore, Cohen's kappa and F1-score were more than 72% on average between all sleep stages. According to the confusion matrix, the algorithm successfully classified sleep sig

关键词： Classification Deep learning Long short-term memory Recurrent neural network Sleep stage scoring Training

来源：评论

学校读者我要写书评

暂无评论

Uncertainty Driven Workflow Scheduling Using Unreliable Cloud Resources

Uncertainty Driven Workflow Scheduling Using Unreliable Clou...

引用

IEEE International Symposium on Network Computing and Applications

作者： Panagiotis Oikonomou Kostas Kolomvatsos Nikos Tziritas Georgios Theodoropoulos Thanasis Loukopoulos Georgios Stamoulis Computer Science and Engineering Southern Univ. of Science and Technology Shenzhen China Computer Science and Telecomm. University of Thessaly Lamia Greece Comp. Science and Biomedical Informatics University of Thessaly Lamia Greece Electrical and Computer Engineering University of Thessaly Volos Greece

ISBN: (数字)9781728183268

ISBN: (纸本)9781728183275

The Cloud infrastructure offers to end users a broad set of heterogenous computational resources using the pay-as-you -go model. These virtualized resources can be provisioned using different pricing models like the unreliable model where resources are provided at a fraction of the cost but with no guarantee for an uninterrupted processing. However, the enormous gamut of opportunities comes with a great caveat as resource management and scheduling decisions are increasingly complicated. Moreover, the presented uncertainty in optimally selecting resources has also a negatively impact on the quality of solutions delivered by scheduling algorithms. In this paper, we present a dynamic scheduling algorithm (i.e., the Uncertainty-Driven Scheduling - UDS algorithm) for the management of scientific workflows in Cloud. Our model minimizes both the makespan and the monetary cost by dynamically selecting reliable or unreliable virtualized resources. For covering the uncertainty in decision making, we adopt a Fuzzy Logic Controller (FLC) to derive the pricing model of the resources that will host every task. We evaluate the performance of the proposed algorithm using real workflow applications being tested under the assumption of different probabilities regarding the revocation of unreliable resources. Numerical results depict the performance of the proposed approach and a comparative assessment reveals the position of the paper in the relevant literature.

关键词： Task analysis Cloud computing Reliability Uncertainty Decision making Computational modeling Quality of service

来源：评论

学校读者我要写书评

暂无评论

Optimized multi-axis spiral projection MR fingerprinting with subspace reconstruction for rapid whole-brain high-isotropic-resolution quantitative imaging

arXiv

引用

arXiv 2021年

作者： Cao, Xiaozhi Liao, Congyu Iyer, Siddharth Srinivasan Wang, Zhixing Zhou, Zihan Dai, Erpeng Liberman, Gilad Dong, Zijing Gong, Ting He, Hongjian Zhong, Jianhui Bilgic, Berkin Setsompop, Kawin Department of Radiology Stanford University StanfordCA United States Department of Electrical Engineering Stanford University StanfordCA United States Athinoula A. Martinos Center for Biomedical Imaging Massachusetts General Hospital CharlestownMA United States Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA United States Department of Biomedical Engineering University of Virginia CharlottesvilleVA United States Center for Brain Imaging Science and Technology College of Biomedical Engineering and Instrumental Science Zhejiang University Zhejiang Hangzhou China Department of Imaging Sciences University of Rochester RochesterNY United States Department of Radiology Harvard Medical School CambridgeMA United States Harvard-MIT Division of Health Sciences and Technology Massachusetts Institute of Technology CambridgeMA United States

Purpose: To improve image quality and accelerate the acquisition of 3D MRF. Methods: Building on the multi-axis spiral-projection MRF technique, a subspace reconstruction with locally low rank (LLR) constraint and a modified spiral-projection spatiotemporal encoding scheme termed tiny-golden-angle-shuffling (TGAS) were implemented for rapid whole-brain high-resolution quantitative mapping. The LLR regularization parameter and the number of subspace bases were tuned using retrospective in-vivo data and simulated examinations, respectively. B0inhomogeneity correction using multi-frequency interpolation was incorporated into the subspace reconstruction to further improve the image quality by mitigating blurring caused by off-resonance effect. Results: The proposed MRF acquisition and reconstruction framework can produce provide high quality 1-mm isotropic whole-brain quantitative maps in a total acquisition time of 1 minute 55 seconds, with higher-quality results than ones obtained from the previous approach in 6 minutes. The comparison of quantitative results indicates that neither the subspace reconstruction nor the TGAS trajectory induce bias for T1 and T2 mapping. High quality whole-brain MRF data were also obtained at 0.66-mm isotropic resolution in 4 minutes using the proposed technique, where the increased resolution was shown to improve visualization of subtle brain structures. Conclusion: The proposed TGAS-SPI-MRF with optimized spiral-projection trajectory and subspace reconstruction can enable high-resolution quantitative mapping with faster acquisition speed. © 2021, CC BY.

关键词： Mapping

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

Background

Objective

Methods

Results

Conclusions

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

Background

Objective

Methods

Results

Conclusions

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：