检索结果-内蒙古大学图书馆

Hyperparameter Optimization Using a Genetic algorithm Considering Verification Time in a Convolutional Neural Network

引用

JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY 2020年第2期15卷 721-726页

作者： Han, Ji-Hoon Choi, Dong-Jin Park, Sang-Uk Hong, Sun-Ki Hoseo Univ Dept Informat Control Engn Asan Chungnam South Korea Hoseo Univ Dept Digital Control Engn Asan Chungnam South Korea

Hyperparameter optimization is a very difficult problem in developing deep learning algorithms. In this paper, a genetic algorithm was applied to solve this problem. The accuracy and the verification time were considered by conducting a fitness evaluation. The algorithm was evaluated by using a simple model that has a single convolution layer and a single fully connected layer. A model with three layers was used. The MNIST dataset and a motor fault diagnosis dataset were used to train the algorithm. The results show that the method is useful for reducing the training time.

关键词： deep learning algorithm Genetic algorithm Hyper-parameter optimization Motor fault diagnosis Convolutional neural network MNIST

来源：评论

学校读者我要写书评

暂无评论

Comparison of the Effects of Mel Coefficients and Spectrogram Images via deep learning in Emotion Classification

引用

TRAITEMENT DU SIGNAL 2020年第1期37卷 51-57页

作者： Demircan, Semiye Ornek, Humar Kahramanli Konya Tech Univ Fac Engn & Nat Sci Dept Comp Engn TR-42075 Konya Turkey Selcuk Univ Fac Technol Dept Comp Engn TR-42075 Konya Turkey

In the present paper, an approach was developed for emotion recognition from speech data using deep learning algorithms, a problem that has gained importance in recent years. Feature extraction manually and feature selection steps were more important in traditional methods for speech emotion recognition. In spite of this, deep learning algorithms were applied to data without any data reduction. The study implemented the triple emotion groups of EmoDB emotion data: Boredom, Neutral, and Sadness-BNS;and Anger, Happiness, and Fear-AHF. Firstly, the spectrogram images resulting from the signal data after preprocessing were classified using AlexNET. Secondly, the results formed from the MelFrequency Cepstrum Coefficients (MFCC) extracted by feature extraction methods to deep Neural Networks (DNN) were compared. The importance and necessity of using manual feature extraction in deep learning was investigated, which remains a very important part of emotion recognition. The experimental results show that emotion recognition through the implementation of the AlexNet architecture to the spectrogram images was more discriminative than that through the implementation of DNN to manually extracted features.

关键词： speech emotion recognition deep Neural Network (DNN) Convolutional Neural Network (CNN) deep learning algorithm Mel-Frequency Cepstrum Coefficients (MFCC)

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of point clouds of building interiors with deep learning: Augmenting training datasets with synthetic BIM-based point clouds

引用

AUTOMATION IN CONSTRUCTION 2020年第0期113卷 103144-000页

作者： Won, Jong Czerniawski, Thomas Leite, Fernanda Univ Texas Austin Dept Civil Architectural & Environm Engn 1 Univ Stn C1752 Austin TX 78712 USA

This paper investigates the viability of using synthetic point clouds generated from building information models (BIMs) to train deep neural networks to perform semantic segmentation of point clouds of building interiors. In order to achieve these goals, this paper first presents a procedure for converting digital 3D BIMs into synthetic point clouds using three commercially available software systems. Then the generated synthetic point clouds are used to train a deep neural network. Semantic segmentation performance is compared for several models trained on: real point clouds, synthetic point clouds, and combinations of real and synthetic point clouds. A key finding is the 7.1% IOU boost in performance achieved when a small real point cloud dataset is augmented by synthetic point clouds for training, as compared to training the classifier on the real data alone. The experimental results confirmed the viability of using synthetic point clouds generated from building information models in combination with small datasets of real point clouds. This opens up the possibility of developing a segmentation model for building interiors that can be applied to as-built modeling of buildings that contain unseen indoor structures.

关键词： Synthetic dataset Point clouds deep learning algorithm Semantic segmentation Building information model

来源：评论

学校读者我要写书评

暂无评论

Multiplex immunofluorescence and single-cell transcriptomic profiling reveal the spatial cell interaction networks in the non-small cell lung cancer microenvironment

引用

CLINICAL AND TRANSLATIONAL MEDICINE 2023年第1期13卷 e1155页

作者： Peng, Haoxin Wu, Xiangrong Liu, Shaopeng He, Miao Xie, Chao Zhong, Ran Liu, Jun Tang, Chenshuo Li, Caichen Xiong, Shan Zheng, Hongbo He, Jianxing Lu, Xu Liang, Wenhua Guangzhou Med Univ Affiliated Hosp 1 Dept Thorac Oncol & Surg China State Key Lab Resp Dis Guangzhou Peoples R China Guangzhou Med Univ Natl Clin Res Ctr Resp Dis Affiliated Hosp 1 Guangzhou Peoples R China Guangzhou Med Univ Nanshan Sch Dept Clin Med Guangzhou Peoples R China Guangdong Polytech Normal Univ Dept Comp Sci Guangzhou 510006 Peoples R China Dept Artificial Intelligence Res Pazhou Lab Guangzhou Peoples R China Genecast Biotechnol Co Ltd Med Dept Beijing Peoples R China First Peoples Hosp Zhaoqing Dept Med Oncol Zhaoqing Peoples R China Guangzhou Med Univ Dept Thorac Surg Affiliated Hosp 1 Guangzhou Peoples R China China State Key Lab Resp Dis Guangzhou Peoples R China Natl Clin Res Ctr Resp Dis Guangzhou Peoples R China Pazhou Lab Guangzhou 510330 Peoples R China

BackgroundConventional immunohistochemistry technologies were limited by the inability to simultaneously detect multiple markers and the lack of identifying spatial relationships among cells, hindering understanding of the biological processes in cancer immunology. MethodsTissue slices of primary tumours from 553 IA similar to IIIB non-small cell lung cancer (NSCLC) cases were stained by multiplex immunofluorescence (mIF) assay for 10 markers, including CD4, CD38, CD20, FOXP3, CD66b, CD8, CD68, PD-L1, CD133 and CD163, evaluating the amounts of 26 phenotypes of cells in tumour nest and tumour stroma. StarDist depth learning model was utilised to determine the spatial location of cells based on mIF graphs. Single-cell RNA sequencing (scRNA-seq) on four primary NSCLC cases was conducted to investigate the putative cell interaction networks. ResultsSpatial proximity among CD20+ B cells, CD4+ T cells and CD38+ T cells (r(2) = 0.41) was observed, whereas the distribution of regulatory T cells was associated with decreased infiltration levels of CD20+ B cells and CD38+ T cells (r(2) = -0.45). Univariate Cox analyses identified closer proximity between CD8+ T cells predicted longer disease-free survival (DFS). In contrast, closer proximity between CD133+ cancer stem cells (CSCs), longer distances between CD4+ T cells and CD20+ B cells, CD4+ T cells and neutrophils, and CD20+ B cells and neutrophils were correlated with dismal DFS. Data from scRNA-seq further showed that spatially adjacent N1-like neutrophils could boost the proliferation and activation of T and B lymphocytes, whereas spatially neighbouring M2-like macrophages showed negative effects. An immune-related risk score (IRRS) system aggregating robust quantitative and spatial prognosticators showed that high-IRRS patients had significantly worse DFS than low-IRRS ones (HR 2.72, 95% CI 1.87-3.94, p < .001). ConclusionsWe developed a framework to analyse the cell interaction networks in tumour microenvironment, reve

关键词： cell interaction networks deep learning algorithm multiplex immunofluorescence single-cell RNA sequencing tumour microenvironment

来源：评论

学校读者我要写书评

暂无评论

High relative amount of nodular calcification in femoral plaques is associated with milder lower extremity arterial disease

引用

BMC CARDIOVASCULAR DISORDERS 2022年第1期22卷 563-563页

作者： Azeez, Mae Laivuori, Mirjami Tolva, Johanna Linder, Nina Lundin, Johan Albaeck, Anders Venermo, Maarit Maeyraenpaeae, Mikko I. Lokki, Marja-Liisa Lokki, A. Inkeri Sinisalo, Juha Univ Helsinki Dept Pathol Transplantat Lab Haartmaninkatu 3 FIN-00014 Helsinki Finland Helsinki Univ Hosp Univ Helsinki Dept Vasc Surg Haartmaninkatu 4 PL 340 FIN-00029 Helsinki Finland Univ Helsinki Inst Mol Med Finland HILIFE Helsinki 00014 Finland Karolinska Inst Dept Publ Hlth Sci Global Hlth IHCAR SE-17177 Stockholm Sweden Univ Helsinki Helsinki Univ Hosp Dept Pathol Haartmaninkatu 3 FIN-00014 Helsinki Finland Univ Helsinki Translat Immunol Res Program Res Programs Unit Haartmaninkatu 8 FIN-00014 Helsinki Finland Helsinki Univ Hosp Heart & Lung Ctr Dept Cardiol PL 340 Haartmaninkatu 4 FIN-00029 Helsinki Finland Univ Helsinki Haartmaninkatu 4 PL 340 FIN-00029 Helsinki Finland

Background: Clinical implications of different types of vascular calcification are poorly understood. The two most abundant forms of calcification, nodular and sheet calcification, have not been quantitatively analyzed in relation to the clinical presentation of lower extremity arterial disease (LEAD).Methods: The study analyzed 51 femoral artery plaques collected during femoral endarterectomy, characterized by the presence of > 90% stenosis. Comprehensive clinical data was obtained from patient records, including magnetic resonance angiography (MRA) images, toe pressure and ankle brachial index measurements and laboratory values. The plaques were longitudinally sectioned, stained with Hematoxylin and Eosin and digitized in a deep learning platform for quantification of the relative area of nodular and sheet calcification to the plaque section area. A deep learning artificial intelligence algorithm was designed and independently validated to reliably quantify nodular calcification and sheet calcification. Vessel measurements and quantity of each calcification category was compared to the risk factors and clinical ***: On average, > 90% stenosed vessels contained 22.4 +/- 12.3% of nodular and 14.5 +/- 11.8% of sheet calcification. Nodular calcification area proportion in lesions with > 90% stenosis is associated with reduced risk of critically low toe pressure (< 30 mmHg) (OR = 0.910, 95% CI = 0.835-0.992, p < 0.05), severely lowered ankle brachial index (< 0.4) (OR = 0.912, 95% CI = 0.84-0.986, p < 0.05), and semi-urgent operation (OR = 0.882, 95% CI = 0.797-0.976, p < 0.05). Sheet calcification did not show any significant ***: Large amount of nodular calcification is associated with less severe LEAD. Patients with nodular calcification may have better flow reserves despite local obstruction.

关键词： Lower extremity arterial disease deep learning algorithm Vascular calcification Sheet calcification Nodular calcification Atherosclerosis

来源：评论

学校读者我要写书评

暂无评论

Predicting Ca²⁺ and Mg²⁺ ligand binding sites by deep neural network algorithm

引用

BMC BIOINFORMATICS 2022年第SUPPL 12期22卷 324-324页

作者： Sun, Kai Hu, Xiuzhen Feng, Zhenxing Wang, Hongbin Lv, Haotian Wang, Ziyang Zhang, Gaimei Xu, Shuang You, Xiaoxiao Inner Mongolia Univ Technol Coll Sci Hohhot 010051 Peoples R China Inner Mongolia Key Lab Stat Anal Theory Life Data Hohhot Peoples R China Inner Mongolia Univ Technol Coll Data Sci & Applicat Hohhot 010051 Peoples R China Hohhot First Hosp Hohhot 010051 Peoples R China

Background Alkaline earth metal ions are important protein binding ligands in human body, and it is of great significance to predict their binding residues. Results In this paper, Mg2+ and Ca2+ ligands are taken as the research objects. Based on the characteristic parameters of protein sequences, amino acids, physicochemical characteristics of amino acids and predicted structural information, deep neural network algorithm is used to predict the binding sites of proteins. By optimizing the hyper-parameters of the deep learning algorithm, the prediction results by the fivefold cross-validation are better than those of the Ionseq method. In addition, to further verify the performance of the proposed model, the undersampling data processing method is adopted, and the prediction results on independent test are better than those obtained by the support vector machine algorithm. Conclusions An efficient method for predicting Mg2+ and Ca2+ ligand binding sites was presented.

关键词： deep learning algorithm Protein Metal ion ligand Binding residue

来源：评论

学校读者我要写书评

暂无评论

A Trainable Open-Source Machine learning AccelerometerActivity Recognition Toolbox: deep learning Approach

引用

JMIR AI 2023年 2卷 e42337页

作者： Wieland, Fluri Nigg, Claudio Univ Bern Inst Sport Sci Dept Hlth Sci Bremgartenstr 145 CH-3012 Bern Switzerland

Background: The accuracy of movement determination software in current activity trackers is insufficient for scientific applications, which are also not open-source. Objective: To address this issue, we developed an accurate, trainable, and open-source smartphone-based activity-tracking toolbox that consists of an Android app (HumanActivityRecorder) and 2 different deep learning algorithms that can be adapted to new behaviors. Methods: We employed a semisupervised deep learning approach to identify the different classes of activity based on accelerometry and gyroscope data, using both our own data and open competition data. Results: Our approach is robust against variation in sampling rate and sensor dimensional input and achieved an accuracy of around 87% in classifying 6 different behaviors on both our own recorded data and the MotionSense data. However, if the dimension-adaptive neural architecture model is tested on our own data, the accuracy drops to 26%, which demonstrates the superiority of our algorithm, which performs at 63% on the MotionSense data used to train the dimension-adaptive neural Conclusions: HumanActivityRecorder is a versatile, retrainable, open-source, and accurate toolbox that is continually tested on new data. This enables researchers to adapt to the behavior being measured and achieve repeatability in scientific studies.

关键词： activity classification deep learning accelerometry open source activity recognition machinelearning activity recorder digital health application smartphone app deep learning algorithm sensor device

来源：评论

学校读者我要写书评

暂无评论

Development of sports industry under the influence of COVID-19 epidemic situation based on big data

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2020年第6期39卷 8867-8875页

作者： Wang Yile Zeng Dashuai Nanyang Normal Univ Sch Phys Educ Nanyang Henan Peoples R China Chengdu Univ Dept Phys Educ Chengdu Sichuan Peoples R China Univ Elect Sci & Technol Chengdu Sichuan Peoples R China

Based on big data, this paper studies the influence of new type of filling pneumonia on the development of sports industry. When selecting the typical economic indicators that reflect the development trend of sports industry, it is found that the data is huge according to the big industrial data, but the information that can be reflected is poor and complex. Therefore, it is necessary to process these big economic data in order to obtain the impact of new coronary pneumonia on the development of sports industry. This paper studies the feature selection algorithm of big data samples, so as to select typical economic indicators from many economic indicators of sports industry to reflect the development trend of sports industry. A deep learning algorithm based on feature selection of big data is proposed. Firstly, a feature selection framework for big data is constructed, and then data fusion and deep learning are carried out. Experiments show that the algorithm can solve the contradiction between large data and poor information. This method has a certain forward-looking, and has a certain reference value for the information discrimination of the development trend of sports industry.

关键词： Sports industry big data industrial development deep learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Assessment of Photovoltaic Capabilities in Urban Environments: A Case Study of Islamabad, Pakistan

引用

JOURNAL OF SOLAR ENERGY ENGINEERING-TRANSACTIONS OF THE ASME 2020年第6期142卷 061006页

作者： Zubair, Muhammad Ghuffar, Sajid Shoaib, Muhammad Awan, Ahmed Bilal Bhatti, Abdul Rauf Majmaah Univ Coll Engn Dept Elect Engn Riyadh 11952 Saudi Arabia Inst Space Technol Dept Space Sci Islamabad 44000 Pakistan Govt Coll Univ Dept Elect Engn Faisalabad 38000 Pakistan

Photovoltaic (PV) estimation in an urban environment requires detection of rooftop area, design of PV system based on optimization on PV placement distance and the study of additional benefit of lower cooling load of building by shading provided by PV panels. The study is aimed at policymakers to introduce renewable energy policy toward net-zero energy buildings in urban areas. In this research, the capital city of Pakistan, Islamabad, is analyzed for rooftop PV capabilities using deep learning algorithms. The area of the rooftop is calculated by extracting buildings in high-resolution satellite imagery using a deep learning algorithm. The site location is analyzed for available solar energy resources. The distance between the rooftop-PV array is optimized based on self-shading losses, coefficient of performance, energy yield, net-zero energy analysis, and reduction of cooling load of the building provided by PV arrays as shading devices. The 40-km(2) area of Islamabad considered in this research can generate 1038 GWh of solar energy annually from its 4.3-km(2) rooftop area by installed capacity of 447 MW PV panels rows placed at 0.75 m apart. The electricity generated by Islamabad can curtail residential load from the national grid and form a near net-zero energy zone while the electrical energy from the grid can be provided to the industries to enhance the economy and reduce unemployment in Pakistan.

关键词： deep learning algorithm photovoltaic arrays energy analysis building energy GIS

来源：评论

学校读者我要写书评

暂无评论

Classification of digital pathological images of non-Hodgkin's lymphoma subtypes based on the fusion of transfer learning and principal component analysis

引用

MEDICAL PHYSICS 2020年第9期47卷 4241-4253页

作者： Zhang, Jianfei Cui, Wensheng Guo, Xiaoyan Wang, Bo Wang, Zhen Qiqihar Univ Sch Comp & Control Engn Qiqihar 161006 Peoples R China

Purpose Non-Hodgkin's lymphoma (NHL) is a serious malignant disease. Delayed diagnosis will cause anemia, increased intracranial pressure, organ failure, and even lead to death. The current main trend in this area is to use deep learning (DL) for disease diagnosis. Extracting classification information from the digital pathology images by DL may realize the automated qualitative and quantitative analysis of NHL. Previously, DL has been used to classify NHL digital pathology images with some success. However, shortcomings still exist in the data preprocessing methods and feature extraction. Therefore, this paper presents a method for the classification of NHL subtypes based on the fusion of transfer learning (TL) and principal component analysis (PCA). Methods First, the NHL digital pathology images were preprocessed by image division and segmentation and then input into the transfer models for fine-tuning and feature extraction. Second, PCA was used to map the extracted features. Finally, a neural network was used as a classifier to classify the mapped features. During the fine-tuning of the transfer models, two methods, freezing all feature extraction layers and fine-tuning all layers, were employed to select the optimal model with the best classification result among all the preselected transfer models. On this basis, the use of freezing the layers' location was discussed and analyzed. Results The results show that the proposed method achieved average fivefold cross-validation accuracies of 100%, 99.73%, and 99.20% for chronic lymphocytic leukemia (CLL), follicular lymphoma (FL), and mantle cell lymphoma (MCL) tumor, and each category has standard deviations 0.00, 0.53, and 0.65, respectively, in the NHL reference dataset. The overall classification accuracy for fivefold cross-validation is 98.93%, which is an increase of 1.26% compared to the latest reported methods, having a lower standard deviation (1.00). Conclusion The method proposed in this paper achieves a

关键词： computer-aided diagnosis (CAD) deep learning algorithm lymphoma principal component analysis (PCA) transfer learning (TL)

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：