检索结果-内蒙古大学图书馆

Conference on Medical Imaging - image-Guided Procedures, Robotic Interventions, and Modeling

作者： Mandal, Shreyasi Bhuiya, Srinjoy Lugez, Elodie Indian Inst Technol Kanpur India Univ Alberta Edmonton AB Canada Toronto Metropolitan Univ Toronto ON Canada

ISBN: (纸本)9781510671614;9781510671607

Purpose: In high-dose-rate prostate brachytherapy, multiple catheters are inserted into the prostate gland. Conventional catheter detection in ultrasound images is hindered by images of a low signal-to-noise ratio, making it difficult to identify the catheters manually. To address this issue, this paper presents an innovative automated deep-learning approach for catheter path reconstruction. Method: A lightweight spatial attention-based autoencoder convolutional neural network is developed to accurately and rapidly segment catheters in real-time volumetric transrectal ultrasound images. To overcome the challenges of noisy and limited annotated data, an auto-augmentation technique is employed, which leverages a controller network to learn the optimal augmentation policy. A 3D random sample consensus-based catheter path reconstruction algorithm is also proposed to transform the catheter segmentations into smooth curves representing their paths. Results: By integrating automated data augmentation with an optimized autoencoder network, structured dropout, and batch normalization techniques, the proposed algorithm successfully detected 98% of the catheter paths tested. The mean tip errors were recorded as 0.18 +/- 0.12 mm, while the mean shaft errors were recorded as 0.39 +/- 0.28 mm, varying depending on the complexity of the catheter curve path. Notably, the proposed algorithm outperformed existing methods by exhibiting faster inference times, with an average inference time of 0.0029 seconds. Conclusion: The proposed methodology offers a comprehensive approach to enhance the accuracy of catheter path reconstruction in prostate brachytherapy. This lightweight neural network also has the potential to significantly improve the prostate brachytherapy workflow by making the catheter reconstruction process time-efficient.

关键词： deep learning Ultrasound Segmentation Convolutional Neural Network High-Dose-Rate Brachytherapy

来源：评论

学校读者我要写书评

暂无评论

Integrated Platform of Underwater Acoustic Signal Simulation and Classification 17

Integrated Platform of Underwater Acoustic Signal Simulation...

引用

17th International Conference on Advanced Computer Theory and Engineering

作者： Zhu, Haoliang Guo, Tiantian Yu, Limin Xian Jiaotong Liverpool Univ Dept Sch Adv Technol Suzhou Peoples R China

ISBN: (纸本)9798331539894;9798331539887

With the increase in marine activities, there is a growing need for efficient and accurate underwater acoustic signal processing techniques. In this study, MATLAB-based graphical user interfaces (GUIs) including Receiver Signal Simulation GUI, Database System GUI and Signal Classification GUI are developed for simulating, sorting, classifying and visualizing underwater acoustic signals with the aim of improving the efficiency and accuracy of acoustic signal processing. By converting one-dimensional acoustic signals into two-dimensional time-frequency spectrogram, this study applies image classification techniques in underwater acoustic signal processing using convolutional neural networks (CNNs), deep neural networks (DNNs), and random forests (RFs) for signal classification. Experimental results show that these machine learning models are able to effectively classify acoustic signals in a low signal-to-noise ratio environment, with CNN and RF showing particular robustness. In addition, the GUI provides a user-friendly platform that greatly simplifies the complex process of analyzing acoustic data. This study not only demonstrates effectiveness of the algorithms in classifying underwater acoustic signals, but also provides new tools and perspectives for marine science research and related application areas. The future work will explore the integration of more machine learning algorithms and extend the functionality of the GUI to support a wider range of acoustic signal processing applications.

关键词： Underwater Acoustic Signal processing Graphical User Interface Machine learning Signal Classification Signal Visualization Geometric Ray Tracing Channel Model

来源：评论

学校读者我要写书评

暂无评论

Bearing Fault Diagnosis Using Machine learning and deep learning Techniques 4th

Bearing Fault Diagnosis Using Machine Learning and Deep Lear...

引用

4th Congress on Intelligent Systems (CIS)

作者： Dhanush, N. Sai Ambika, P. S. Amrita Vishwa Vidyapeetham Amrita Sch Artificial Intelligence Ctr Computat Engn & Networking CEN Coimbatore India

ISBN: (纸本)9789819990368;9789819990375

Machine Fault diagnosis plays a vital role in ensuring the reliability and efficiency of industrial systems. Among various components, roller element bearing is prone to failures due to critical function in supporting rotating machinery. This paper proposes a machine fault diagnosis approach specifically tailored for roller element bearings. The methodology combines vibration analysis, signal processing techniques, and machine learning algorithms to accurately classify bearing faults. Firstly, vibration signals are acquired from the machine using accelerometers, and relevant features are extracted using time-domain, frequency-domain, and statistical methods. Subsequently, autoencoders are also used to extract more features with the aid of existing features. Finally, few state-of-the-art machine learning algorithms such as support vector machines, random forests are trained to classify the fault types. Experimental results on a simulated dataset and real-world scenarios illustrate effectiveness and accuracy of proposed approach in diagnosing rolling bearing faults. This developed methodology offers a practical solution for conditional monitoring and predictive maintenance, enabling timely detection and mitigation of bearing faults, thereby enhancing system reliability, minimizing downtime, and reducing maintenance costs.

关键词： Machine fault diagnosis Autoencoder Machine learning deep learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Reinforcement learning-Based Behavior Pattern Analytics Model for Action and learning-Plan Recommendation in Children 2

A Novel Reinforcement Learning-Based Behavior Pattern Analyt...

引用

2nd IEEE International Conference on Data Science and Network Security, ICDSNS 2024

作者： Chaturved, E. Samatha Sree Gladence, L. Mary Tamilnadu India

ISBN: (纸本)9798350373110

'Virtual reality (VR), Augmented reality (AR), and Mixed reality (MR)' are the emerging approaches to enhance the real-time interaction and user perception between real and virtual elements. The critical issues occurred in the deep learning networks due to the nature of black-box and limited labeled data. ABA is an effective approach to children's learning needs. The explainable artificial intelligence-related augmented learning with the support of applied behavior analytics mechanisms can assist parents and teachers by analyzing facial expressions, movements, speech, and other data to provide personalized action recommendations and learning plans for children. So, in this paper, an enhanced reinforcement learning-based recommendation system is developed. The required data for the development of the suggested model is gathered from the EEG signals taken from the children, their language, speech signals, and images. The features from these data are retrieved. From the EEG signal, the signal features are extracted. From the language input, the features are extracted with the help of the Term Frequency-Inverse Document Frequency (TF-IDF) approach. The cepstral and the spectral features are extracted from the speech signals. The feature from the image input is extracted via Convolutional Neural Network (CNN). Once all the features are extracted, then the recommendation output is obtained using these features as the input. The extracted features are given as input to the reinforcement learning model. From the reinforcement learning model, the recommendation outcome is generated. Experimentations are performed to determine the effectiveness of the suggested model by validating it with conventional recommendation frameworks. © 2024 IEEE.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Machine learning and deep learning Techniques for Hand Gesture Recognition Using Surface Electromyography 3

Comparative Analysis of Machine Learning and Deep Learning T...

引用

3rd International Congress of Biomedical Engineering and Bioengineering, CIIBBI 2024

作者： Hernandez Salazar, Sara Valentina Preciado, Javier Chaparro Munoz, Santiago Agudo Escuela Colombiana de Lngeniería Maestría en Ingeniería Electrónica Julio Garavito Bogotá Colombia

ISBN: (纸本)9798331532352

Hand gesture recognition for the control of rehabilitation and assistive technologies is still a field to be explored. This study investigated the effectiveness of different methods for classifying hand gestures using surface electromyography (sEMG) signals collected from the upper Thalmic Myo armband in experiment two of the Ninapro 5 database. Initially, traditional Machine learning (ML) techniques were applied, an extraction of temporary characteristics was made, and dimensionality reduction through Principal Component Analysis (PCA) and the Kruskal-Wallis test, followed by four classical classifiers, Decision Trees (DT), K-Nearest Neighbor (KNN), Bayes-Gaussian (NBG) and Support Vector Machine (SVM). However, the results were unsatisfactory, with a maximum accuracy of only 0.10. Subsequently, deep learning (DL) techniques were adopted, where the sEMG signals were transformed into images through the Fourier Transform, and although they showed a significant improvement, with a training accuracy of approximately 0.90, the accuracy in validation and testing remained around 0.50. This indicates possible overtraining issues and a lack of differentiation between classes of images. Despite these challenges, DL proved superior to ML in classifying hand gestures from sEMG signals. It is concluded that further research and improvement of signal-to-image transformation techniques are needed to achieve more accurate and feasible classification in real-time applications. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Study on CPU and FPGA in Artificial Intelligence and Machine learning 13

Study on CPU and FPGA in Artificial Intelligence and Machine...

引用

13th International Conference of Information and Communication Technology, ICTech 2024

作者： Zhou, Peng School of Electrical and Computer Engineering Nang Fang College Guangzhou China

ISBN: (纸本)9798350376258

—This article provides a comprehensive examination of the roles played by Central processing Units (CPUs) and Field-Programmable Gate Arrays (FPGAs) in the realms of Artificial Intelligence (AI) and Machine learning (ML), with a particular emphasis on the superiorities of FPGA over CPU in certain aspects. It is observed that CPUs, while versatile, fall short in aspects such as power efficiency, handling the intricacies of deep learning models, and performing large-scale matrix computations. FPGAs, on the other hand, are lauded for their remarkable flexibility, low latency, superior energy efficiency, and inherent parallelism — qualities that are immensely beneficial in AI and ML applications. The article delves into the real-world applications of both CPUs and FPGAs, showcasing specific scenarios where each excels. For instance, the low latency advantage of FPGAs is crucial in image processing tasks, where quick, efficient processing can be pivotal. In contrast, CPUs are better suited for smaller-scale or general-purpose computing tasks where flexibility and ease of programming are key. This contrast highlights the nuanced roles that these technologies play in the broader context of AI and ML. Further, the article contrasts CPUs and FPGAs in terms of architecture, design philosophy, and suitability for various AI and ML workloads. While CPUs offer a general-purpose solution with a well-established ecosystem, FPGAs provide a customizable platform that can be tailored to specific needs, offering a more efficient solution for certain types of computational tasks in AI and ML. This comparison not only sheds light on the inherent strengths and weaknesses of each technology but also underscores the importance of choosing the right tool for specific AI and ML applications, thereby optimizing performance, efficiency, and power consumption. ©2024 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Unidirectional Encrypted Traffic Classification: A Survey 5

Unidirectional Encrypted Traffic Classification: A Survey

引用

5th Asia-Pacific Conference on image processing, Electronics and Computers, IPEC 2024

作者： Yang, Chen Gu, Zheyuan Wei, Yuhao Xiong, Gang Gou, Gaopeng Yao, Shan Yu, Yang Institute of Information Engineering University of Chinese Academy of Sciences Chinese Academy of Sciences School of Cyber Security Beijing China 360 Digital Security Technology Group Co. Ltd Beijing China

ISBN: (纸本)9798350374407

Techniques for classifying encrypted traffic typically rely on learning the interaction patterns between communicating parties to determine the traffic category. However, with the widespread application of asymmetric routing mechanisms and link load balancing, a large volume of network equipment cannot observe the complete bidirectional traffic, posing challenges to the deployment of existing methods. In order to solve this problem, research into unidirectional encrypted traffic classification methods has gradually emerged. In this paper, we provide a comprehensive survey of existing unidirectional encrypted traffic classification methods and showcase the latest research progress in this field. Based on how to resolve the contradiction between unidirectional traffic data and model reliance on bidirectional interaction features, we categorize existing methods into three main types: 1) unidirectional feature separation-based methods, 2) unidirectional feature mining-based methods, and 3) pairwise feature prediction-based methods. This paper details the characteristics of each method and discusses their advantages and disadvantages. We hope our work will draw researchers' attention to the real-world network environment, thereby better promoting the development of the encrypted traffic classification field. © 2024 IEEE.

关键词： deep learning

来源：评论

学校读者我要写书评

暂无评论

1D in-situ convolution system based on vibration signal for real-time structural health monitoring

引用

NANO ENERGY 2024年 127卷

作者： Fan, Yuyang Zhang, Xianghong Cheng, Enping Qin, Congyao Qin, Ningpu Wu, Jianxin Guo, Xiao Chen, Huipeng Guo, Tailiang Fuzhou Univ Inst Optoelect Display Natl & Local United Engn Lab Flat Panel Display Te Fuzhou 350002 Peoples R China Fujian Sci & Technol Innovat Lab Optoelect Informa Fuzhou 350100 Peoples R China Fuzhou Univ Sch Architecture & Urban Rural Planning Fuzhou 350002 Peoples R China

Structural health monitoring (SHM) of bridges and other building structures has been widely concerned by researchers and engineers. SHM achieved a breakthrough by using convolutional neural networks (CNNs) in deep learning. However, the traditional vision based SHM can not monitor the invisible structural damage. The traditional vibration based SHM needs to convert 1D signals into 2D images, which is easy to lose information in the conversion process, and requires a large scale of hardware circuit. In this paper, a 1D in-situ convolution system (ICS) based on vibration signal is proposed. The SHM is realized by 1D convolution calculation of the structural vibration signal detected by a high-sensitivity triboelectric nanogenerator (TENG) through one in-situ convolution transistor. Based on the natural adaptability of TENG to 1D convolution calculation, and combined with the ability of a single in-situ convolution device to replace more than eight multi conductance devices, ICS can reduce the required SHM computing array area by over 87.5%. Finally, the real-time identification of different positions and different damage levels of the steel bridge was realized by simulating the vibration response of the steel bridge.

关键词： Convolution transistor Triboelectric nanogenerator Structure health monitoring 1D convolution calculation Vibration signal processing

来源：评论

学校读者我要写书评

暂无评论

image Reconstruction Algorithm in Electrical Impedance Tomography Based on Improved CNN-RBF Model 12th

Image Reconstruction Algorithm in Electrical Impedance Tomog...

引用

12th IFMBE Asian-Pacific Conference on Medical and Biological Engineering (APCMBE)

作者： Zhang, Liyuan Liu, Xuechao Li, Lei Fu, Feng Jin, Li Yang, Bin Air Force Med Univ Dept Biomed Engn Xian 710032 Peoples R China Xi An Jiao Tong Univ Sch Software Engn Xian Peoples R China

ISBN: (纸本)9783031514845;9783031514852

Electrical Impedance Tomography (EIT) is a noninvasive and real-time medical imaging technique, which means more potential applications for clinical diagnosis and treatments. However, EIT reconstruction problem is a highly nonlinear and ill-posed problem which will lead to poor reconstruction quality. Since neural networks have been proven to fit any nonlinearmapping theoretically, there are significant advantages of neural networks for EIT reconstruction. Convolutional neural networks (CNN), as one of the most famous neural networks, have powerful spatial feature extraction capabilities. Radial basis function (RBF) neural networks represent local approximators to nonlinear mapping, which has the significant advantage of short training time. In this paper, we proposed a two-part solver for the EIT reconstruction problem based on deep learning. RBF neural networks are used to solve the EIT reconstruction problem while CNN are used for feature extraction.

关键词： deep learning Electrical Impedance Tomography image Reconstruction

来源：评论

学校读者我要写书评

暂无评论

Vision Transformer Model for Predicting the Severity of Diabetic Retinopathy in Fundus Photography-Based Retina images

引用

IEEE ACCESS 2023年 11卷 117546-117561页

作者： Nazih, Waleed Aseeri, Ahmad O. Atallah, Osama Youssef El-Sappagh, Shaker Prince Sattam Bin Abdulaziz Univ Coll Comp Engn & Sci Dept Comp Sci Al Kharj 11942 Saudi Arabia Cairo Higher Inst Engn Comp Sci & Management Dept Comp Sci Cairo 11865 Egypt Univ Alexandria Med Res Inst Dept Biomed Engn Alexandria 21561 Egypt Galala Univ Fac Comp Sci & Engn Suez 435611 Egypt Benha Univ Fac Comp & Artificial Intelligence Informat Syst Dept Banha 13518 Egypt

Diabetic Retinopathy (DR) is a result of prolonged diabetes with poor blood sugar management. It causes vision problems and blindness due to the deformation of the human retina. Recently, DR has become a crucial medical problem that affects the health and life of people. Diagnosis of DR can be done manually by ophthalmologists, but this is cumbersome and time consuming especially in the current overloaded physician's environment. The early detection and prevention of DR, a severe complication of diabetes that can lead to blindness, require an automatic, accurate, and personalized machine learning-based method. Various deep learning algorithms, particularly convolutional neural networks (CNNs), have been investigated for detecting different stages of DR. Recently, transformers have proved their capabilities in natural language processing. Vision transformers (ViTs) are extensions of these models to capture long-range dependencies in images, which achieved better results than CNN models. However, ViT always needs huge datasets to learn properly, and this condition reduced its applicability in DR domain. Recently, a new real-world and large fundus image dataset called fine-grained annotated diabetic retinopathy (FGADR) has been released which supported the application of ViT in DR diagnosis domain. The literature has not explored FGADR to optimize ViT models. In this paper, we propose a novel ViT based deep learning pipeline for detecting the severity stages of DR based on fundus photography-based retina images. The model has been built using FGADR dataset. The model has been optimized using a new optimizer called AdamW to detect the global context of images. Because FGADR is an imbalanced dataset, we combine several techniques for handling this issue including the usage of F1-score as the optimization metric, data augmentation, class weights, label smoothing, and focal loss. Extensive experiments have been conducted to explore the role of ViT with different data balan

关键词： Retina Transformers Feature extraction Diabetic retinopathy Convolutional neural networks Medical services Blindness deep learning Machine learning vision transformer diabetic retinopathy machine learning disease diagnosis

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：