检索结果-内蒙古大学图书馆

3d matters! 3d-RISM and 3d convolutional neural network for accurate bioaccumulation prediction

JOURNAL OF PHYSICS-CONdENSEd MATTER 2018年第32期30卷 32LT03-32LT03页

作者： Sosnin, Sergey Misin, Maksim Palmer, david S. Fedorov, Maxim V. Skolkovo Inst Sci & Technol Ctr Computat & Data Intens Sci & Engn Nobelya Ulitsa 3 Moscow 121205 Russia Univ Tartu Inst Chem Ravila 14a EE-50411 Tartu Estonia Univ Strathclyde Dept Pure & Appl Chem 295 Cathedral St Glasgow G1 1XL Lanark Scotland Univ Strathclyde SUPA Dept Phys Glasgow Lanark Scotland

In this work, we present a new method for predicting complex physical-chemical properties of organic molecules. The approach utilizes 3d convolutional neural network (ActivNet4) that uses solvent spatial distributions around solutes as input. These spatial distributions are obtained by a molecular theory called three-dimensional reference interaction site model. We have shown that the method allows one to achieve a good accuracy of prediction of bioconcentration factor which is difficult to predict by direct application of methods of molecular theory or simulations. Our research demonstrates that combination of molecular theories with modern machine learning approaches can be effectively used for predicting properties that are otherwise inaccessible to purely theory-based models.

关键词： 3d RISM 3d convolutional neural network molecular theory of liquids

来源：评论

学校读者我要写书评

暂无评论

FractureNet:A 3d convolutional neural network Based on the Architecture of m-Ary Tree for Fracture Type Identification

引用

IEEE TRANSACTIONS ON MEdICAL IMAGING 2022年第5期41卷 1196-1207页

作者： Chen, Yiren He, Kunjin Hao, Bo Weng, Yiping Chen, Zhengming Hohai Univ Informat Dept Nanjing 211100 Peoples R China Anqing Normal Univ Sch Comp & Informat Anqing 246133 Peoples R China Changzhou City Key Lab Orthoped Implants Digital Changzhou 213022 Peoples R China Changzhou 2 Peoples Hosp Dept Orthoped Changzhou 213003 Peoples R China

To address the problem of automatic identification of fine-grained fracture types, in this paper, we propose a novel framework using 3d convolutional neural network (CNN) to learn fracture features from voxelized bone models which are obtained by establishing isomorphic mapping from fractured bones to a voxelized template. The network, which is named FractureNet, consists of four discriminators forming a multi-stage hierarchy. Each discriminator includes multiple sub-classifiers. These sub-classifiers are chained by two kinds of feature chains (feature map chain and classification feature chain) in the form of a full m-ary tree to perform multi-stage classification tasks. The features learned and classification results obtained at previous stages serve as prior knowledge for current learning and classification. All sub-classifiers are jointly learned in an end-to-end network via a multi-stage loss function integrating losses of the four discriminators. To make our FractureNet more robust and accurate, a data augmentation strategy termed r-combination with constraints is further proposed on the basis of an adjacency relation and a continuity relation between voxels to create a large-scale fracture dataset of voxel models. Extensive experiments show that the proposed method can recognize various fracture types in patients accurately and effectively, and enables significant improvements over the state-of-thearts on a variety of fracture recognition tasks. Moreover, ancillary experiments on the CIFAR-10 and the PadChest datasets at large scales further support the superior performance of the proposed FractureNet.

关键词： Fracture type identification 3d convolutional neural network feature chaining hierarchical cascade network constrained r-combination strategy

来源：评论

学校读者我要写书评

暂无评论

The OCdA-Net: a 3d convolutional neural network-based system for classification and staging of ovarian cancer patients using [¹⁸F]FdG PET/CT examinations

引用

ANNALS OF NUCLEAR MEdICINE 2023年第12期37卷 645-654页

作者： Sadeghi, Mohammad Hossein Sina, Sedigheh Alavi, Mehrosadat Giammarile, Francesco Shiraz Univ Sch Mech Engn Nucl Engn Dept Shiraz Iran Shiraz Univ Radiat Res Ctr Sch Mech Engn Shiraz Iran Shiraz Univ Med Sci Dept Nucl Med Shiraz Iran IAEA Div Human Hlth Nucl Med & Diagnost Imaging Sect Vienna Austria

Objective To create the 3d convolutional neural network (CNN)-based system that can use whole-body [F-18]FdG PET for recurrence/post-therapy surveillance in ovarian cancer (OC).Methods In this study, 1224 image sets from OC patients who underwent whole-body [F-18]FdG PET/CT at Kowsar Hospital between April 2019 and May 2022 were investigated. For recurrence/post-therapy surveillance, diagnostic classification as cancerous, and non-cancerous and staging as stage III, and stage IV were determined by pathological diagnosis and specialists' interpretation. New deep neural network algorithms, the OCdAc-Net, and the OCdAs-Net were developed for diagnostic classification and staging of OC patients using [F-18]FdG PET/CT images. Examinations were divided into independent training (75%), validation (10%), and testing (15%) *** This study included 37 women (mean age 56.3 years;age range 36-83 years). data augmentation techniques were applied to the images in two phases. There were 1224 image sets for diagnostic classification and staging. For the test set, 170 image sets were considered for diagnostic classification and staging. The OCdAc-Net areas under the receiver operating characteristic curve (AUCs) and overall accuracy for diagnostic classification were 0.990 and 0.92, respectively. The OCdAs-Net achieved areas under the receiver operating characteristic curve (AUCs) of 0.995 and overall accuracy of 0.94 for *** The proposed 3d CNN-based models provide potential tools for recurrence/post-therapy surveillance in OC. The OCdAc-Net and the OCdAs-Net model provide a new prognostic analysis method that can utilize PET images without pathological findings for diagnostic classification and staging.

关键词： 3d convolutional neural network Classification Staging Ovarian cancer PET/CT

来源：评论

学校读者我要写书评

暂无评论

3d convolutional neural network for Low-Light Image Sequence Enhancement in SLAM

引用

REMOTE SENSING 2022年第16期14卷 3985-3985页

作者： Quan, Yizhuo Fu, dong Chang, Yuanfei Wang, Chengbo Chinese Acad Sci Aerosp Informat Reaserch Inst 9 Dengzhuang South Rd Beijing 100094 Peoples R China Univ Chinese Acad Sci 19 A Yuquan Rd Beijing 100049 Peoples R China

Typical visual simultaneous localization and mapping (SLAM) systems rely on front-end odometry for feature extraction and matching to establish the relations between adjacent images. In a low-light environment, the image obtained by a camera is dim and shows scarce information, hindering the extraction of sufficient stable feature points, consequently undermining visual SLAM. Most existing methods focus on low-light enhancement of a single image, neglecting the strong temporal correlation across images in visual SLAM. We propose a method that leverages the temporal information of an input image sequence to enhance the low-light image and employed the enhanced result to improve the feature extraction and matching quality of visual SLAM. Our method trains a three-dimensional convolutional neural network to estimate pixelwise grayscale transformation curves to obtain a low-light enhancement image. Then, the grayscale transformation curves are iteratively applied to obtain the final enhanced result. The training process of the network does not require any paired reference images. We also introduced a spatial consistency loss for the enhanced image to retain the content and texture of the original image. We further integrated our method into VINS-Mono and compared with similar low-light image enhancement methods on the TUM-VI public dataset. The proposed method provides a lower positioning error. The positioning root-mean-squared error of our method is 19.83% lower than that of Zero-dCE++ in low-light environments. Moreover, the proposed network achieves real-time operation, being suitable for integration into a SLAM system.

关键词： visual inertial system low-light image sequence enhancement 3d convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Rapid estimation of permeability from digital rock using 3d convolutional neural network

引用

COMPUTATIONAL GEOSCIENCES 2020年第4期24卷 1523-1539页

作者： Hong, Jin Liu, Jie Sun Yat Sen Univ Sch Earth Sci & Engn Guangzhou 510275 Peoples R China

Permeability and its anisotropy are of central importance for groundwater and hydrocarbon migration. Existing fluid dynamics methods for computing permeability have common shortcomings, i.e., high computational complexity and long computational time, reducing the potential of these methods in practical applications. In view of this, a 3d CNN-based approach for rapidly estimating permeability in anisotropic rock is proposed. Using high-resolution X-ray microtomographic images of a sandstone sample, numerous samples of the size of 100-cube voxels were generated firstly by a series of image manipulation techniques. The shrinking and expanding algorithms are employed as the data augmentation methods to strengthen the role of porosity and specific surface area (SSA) since these two parameters are critical to estimate permeability. Afterwards, direct pore-scale modeling with Lattice-Boltzmann method (LBM) was utilized to compute the permeabilities in the direction of three coordinate axes and mean permeability as the ground truth. A dataset including 3158 samples for training and 57 samples for testing were obtained. Four 3d CNN models with the same network structure, corresponding to permeabilities in 3 directions and in average, were built and trained. Based on those trained models, the satisfactory predictions of the permeabilities inx-,y-, andz-axis directions and the mean permeability were achieved withR(2)scores of 0.8972, 0.8821, 0.8201, and 0.9155, respectively. Furthermore, those proposed 3d CNN models achieved good generalization ability in predicting the permeability of other samples. The trained model takes only tens of milliseconds on average to predict the permeability of one sample in one axial direction, about 10,000 times faster than LBM. The promising performance clearly demonstrates the effectiveness of 3d CNN-based approach in rapidly estimating permeability in anisotropic rock. This new approach provides an alternative way to calculate permeability wi

关键词： Anisotropic rock digital rock physics Permeability estimation 3d convolutional neural network Shrinking and expanding

来源：评论

学校读者我要写书评

暂无评论

Organ segmentation from computed tomography images using the 3d convolutional neural network: a systematic review

引用

INTERNATIONAL JOURNAL OF MULTIMEdIA INFORMATION RETRIEVAL 2022年第3期11卷 315-331页

作者： Ilesanmi, Ademola E. Ilesanmi, Taiwo Idowu, Oluwagbenga P. Torigian, drew A. Udupa, Jayaram K. Univ Penn Dept Radiol Med Image Proc Grp Goddard Bldg3710 Hamilton Walk6th Floor Philadelphia PA 19104 USA Natl Populat Commiss Abuja Nigeria Sichuan Univ West China Hosp Chengdu Peoples R China

Computed tomography images are scans that combine a series of X-rays with computer processing techniques to display organs in the body. Recently, 3d CNN models have become effective in tasks relating to recognition, delineation, and classification. Therefore we propose a review to summarize different 3d CNN algorithms for segmenting organs in computed tomography images. This work systematically applies a two-stage procedure for review. A thorough screening of abstracts and titles to ascertain their relevance was done. Research papers published in the academic repositories were selected, analyzed, and reviewed. Insight relating to 3d organ segmentation is provided, with content such as database usage, disadvantages, and advantages. A comparison of two accuracies was carried out with a graph depicting database categories. Important insights, limitations, observations, and future directions were elucidated. After careful investigation, we observe that the encoder-decoder network is predominant for segmentation. The encoder-decoder framework provides a seamless procedure to segment CT images. A prediction of future trends with insightful recommendations for researchers is proposed. Finally, findings suggest that CNN algorithms produce good accuracies despite their limitations.

关键词： 3d convolutional neural network Computed tomography images Segmentation of organs Image preprocessing

来源：评论

学校读者我要写书评

暂无评论

Urban dynamic Logistics Pattern Mining with 3d convolutional neural network

Urban Dynamic Logistics Pattern Mining with 3D Convolutional...

引用

IEEE International Conference on Big data (Big data)

作者： Wen, Rong yan, Wenjing Singapore Inst Mfg Technol Planning & Operat Management Grp 2 Fusionopolis Way Singapore Singapore

ISBN: (纸本)9781538650356

With wide applications of various types of sensors and web-based software, massive location and time related logistics data is available. The massive spatial and temporal data significantly enhances visibility of modern logistics activities, however, integrating continuous time information to explore spatio-temporal pattern is an outstanding challenge. In this research, we develop a data mining method using image-based method to convert massive spatio-temporal information into heatmap based image sequence where spatial data can be embedded into image frames while image sequence represent time domain. By developing architecture of deep learning neural network with three-dimensional (3d) kernels for different convolution layers, 3d convolutional computation can be applied to images with certain temporal depth to extract dynamic spatio-temporal patterns. The method was validated with real taxi data of New York City. Experimental results demonstrated that dynamic logistics patterns could be identified with historical spatio-temporal information. Comparison of clusters generated by different learning methods showed that the proposed method could produce more accurate spatial temporal pattern.

关键词： Spatio-temporal data mining deep learning 3d convolutional neural network dynamic pattern mining logistics and transportation

来源：评论

学校读者我要写书评

暂无评论

Abnormal behavior recognition based on edge feature and 3d convolutional neural network 35

Abnormal behavior recognition based on edge feature and 3D c...

引用

35th Youth Academic Annual Conference of Chinese-Association-of-Automation (YAC)

作者： Bian, Chunlei Xu, Yiming Wang, Li Gu, Haifeng Zhou, Fangjie Nantong Univ Coll Elect Engn Nantong Peoples R China

ISBN: (纸本)9781728176840

with the rapid development of artificial intelligence and 5G technology, the development process of smart city has been greatly accelerated. An important part of the development of smart city is to improve the ability of video monitoring equipment to analyze the content of the video. This paper is to identify the specific abnormal behaviors in the video, including climbing, fighting and falling. The identification of these abnormal behaviors is helpful to find out the danger in time and intervene to ensure personal safety and social stability. Therefore, this paper proposes a deep learning method based on edge extraction. In the first step, the video is decomposed into several frames, and the canny edge detection algorithm is used to extract the edge of the images. In the second step, the extracted edge information is used as the input of the 3d convolutional neural network, and the 3d convolutional neural network model is trained iteratively to recognize the abnormal behaviors. Through a large number of experiments, the effectiveness of the proposed algorithm is verified.

关键词： behavior recognition 3d convolutional neural network deep learning edge extraction

来源：评论

学校读者我要写书评

暂无评论

depthParkNet: A 3d convolutional neural network with depth-Aware Coordinate Attention for PET-Based Parkinson's disease diagnosis 20th

DepthParkNet: A 3D Convolutional Neural Network with Depth-A...

引用

20th International Conference on Intelligent Computing (ICIC)

作者： Li, Maoyuan Chen, Ling Chu, Jianmin Shi, Xinchong Zhang, Xiangsong Zhao, Gansen Tang, Hua South China Normal Univ Sch Comp Sci Guangzhou Peoples R China Key Lab Cloud Secur & Assessment Technol Guangzho Guangzhou Peoples R China Sun Yat Sen Univ Affiliated Hosp 1 Dept Neurol Guangzhou Peoples R China Sun Yat Sen Univ Affiliated Hosp 1 Dept Nucl Med Guangzhou Peoples R China

ISBN: (纸本)9789819756889;9789819756896

Positron Emission Tomography (PET) imaging is essential for accurately diagnosing Parkinson's disease (Pd). With the advancement of deep convolutional networks, 3d convolutional neural networks (CNNs) have been extensively utilized to analyze PET images. However, the limited availability of PET imaging data represents a significant bottleneck in this field. Training on small datasets could restrict the performance of 3d CNNs. Moreover, models trained on small datasets tend to exhibit weak generalizability, making it challenging to handle variations in PET images caused by clinical factors. This paper proposes depth-aware coordinate attention that integrates clinical prior knowledge. It's composed of two modules: the coordinate attention module and the depth-aware attention module. The coordinate attention module could preserve positional information while capturing long-range dependencies. Meanwhile, the depth-aware attention module aims to reveal the relationships among various depths within a PET image. These innovations enable our depthParkNet to extract critical features from limited PET imaging data, thus enhancing its performance. Additionally, this paper proposes an augmentation pipeline named Pdaug, which includes three transformations aimed at improving the generalizability of the model. The proposed model is validated on two datasets, demonstrating outstanding performance with balanced accuracy of 99.17% and 98.13%.

关键词： Parkinson's disease deep learning 3d convolutional neural network Attention mechanism Class imbalance

来源：评论

学校读者我要写书评

暂无评论

Smoke Video detection Algorithm Based On 3d convolutional neural network 34

Smoke Video Detection Algorithm Based On 3D Convolutional Ne...

引用

34th Chinese Control and decision Conference (CCdC)

作者： Shi, Zhen Sun, Rui Huo, Mingge Harbin Engn Univ Inst Intelligent Syst Sci & Engn Harbin 150001 Peoples R China Beijing Electromech Engn Res Inst Beijing Beijing Peoples R China

ISBN: (纸本)9781665478960

Smoke detection can play a key role in the early stage of a fire. Traditional video smoke detection mainly extracts image features of smoke and uses machine learning methods to classify and recognize smoke. However, smoke video is an image of continuous frames, and adjacent frames of images are related to each other. If time complexity is ignored and only spatial features are considered, the key features cannot be fully extracted. By combining the 3d convolutional neural network and introducing the attention mechanism, the temporal and spatial features of the smoke can be effectively weighted and merged, and the 3x1x1 and 1x3x3 convolution kernels are used instead of the 3x3x3 convolution kernels. Reduce model parameters. In order to eliminate most of the non-smoke areas to reduce the time complexity of the system, the MobileNetV2-SSd algorithm is used to do a priori score to locate the suspected smoke areas. Experimental results show that compared to traditional smoke detection methods and 2d smoke detection convolutional networks, the accuracy of smoke detection has been improved.

关键词： Smoke video detection 3d convolutional neural network Temporal and spatial characteristics Attention mechanism Mobilenetv2-SSd

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：