检索结果-内蒙古大学图书馆

A deep-learning algorithm (AIFORIA) for classification of hematopoietic cells in bone marrow aspirate smears based on nine cell classes-a feasible approach for routine screening?

引用

JOURNAL OF HEMATOPATHOLOGY 2025年第1期18卷 1-12页

作者： Saft, Leonie Vaara, Emma Ljung, Elin Kwiecinska, Anna Kumar, Darshan Timar, Botond Karolinska Univ Hosp Clin Pathol & Canc Diagnost Stockholm Sweden Karolinska Inst Dept Oncol Pathol Stockholm Sweden Aifor Technol Plc Helsinki Finland Semmelweis Univ Dept Pathol & Expt Canc Res 1 Budapest Hungary

Bone marrow cytology plays a key role for the diagnosis and classification of hematological disease and is often the first step in the acute setting of unclear cytopenia. AI applications represent a powerful tool in digital image analysis and can improve the diagnostic workflow and accuracy. The aim of this study was to develop an algorithm for the automated detection and classification of hematopoietic cells in digitized bone marrow aspirate smears for potential implementation in the clinical laboratory. The AIFORIA create platform (Aiforia Technologies, Plc, Helsinki, Finland) was used to develop a convolutional neural network algorithm based on nine cell classes. Digitized bone marrow aspirate smears from normal hospital controls were used for AI training. External validation was performed on separate data sets. Automated cell classification was assessed in whole-slide images (WSI) and regions of interest (ROI). A total of 1950 single-cell annotations were applied for AI training with a final total class error of 0.15% with 99.9% precision and sensitivity (FI-score 99.2%). External validation showed an overall precision and sensitivity of 96% and 97% and a F1-score of 96%. Automated cell classification correlated highly across ROI with variable correlation to WSI. The average execution time for classifying 500 hematopoietic cells was < 1 s and <= 260 s for WSI. A cloud-based, deep-learning algorithm for automated detection and classification of hematopoietic cells in bone marrow aspirate smears is a very useful, reliable, and rapid screening tool in combination with cytomorphology.

关键词： deep-learning algorithm Hematopoietic cells Bone marrow

来源：评论

学校读者我要写书评

暂无评论

Development and validation of a deep-learning algorithm for the detection of neovascular age-related macular degeneration from colour fundus photographs

引用

CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY 2019年第8期47卷 1009-1018页

作者： Keel, Stuart Li, Zhixi Scheetz, Jane Robman, Liubov Phung, James Makeyeva, Galina Aung, KhinZaw Liu, Chi Yan, Xixi Meng, Wei Guymer, Robyn Chang, Robert He, Mingguang Univ Melbourne Royal Victorian Eye & Ear Hosp Ctr Eye Res Australia Melbourne Vic Australia Sun Yat Sen Univ Zhongshan Ophthalm Ctr State Key Lab Ophthalmol Guangzhou Guangdong Peoples R China Monash Univ Melbourne Melbourne Vic Australia Healgoo Interact Med Technol Co Ltd Guangzhou Guangdong Peoples R China Stanford Univ Byers Eye Inst Dept Ophthalmol Palo Alto CA 94304 USA

Importance Detection of early onset neovascular age-related macular degeneration (AMD) is critical to protecting vision. Background To describe the development and validation of a deep-learning algorithm (DLA) for the detection of neovascular age-related macular degeneration. Design Development and validation of a DLA using retrospective datasets. Participants We developed and trained the DLA using 56 113 retinal images and an additional 86 162 images from an independent dataset to externally validate the DLA. All images were non-stereoscopic and retrospectively collected. Methods The internal validation dataset was derived from real-world clinical settings in China. Gold standard grading was assigned when consensus was reached by three individual ophthalmologists. The DLA classified 31 247 images as gradable and 24 866 as ungradable (poor quality or poor field definition). These ungradable images were used to create a classification model for image quality. Efficiency and diagnostic accuracy were tested using 86 162 images derived from the Melbourne Collaborative Cohort Study. Neovascular AMD and/or ungradable outcome in one or both eyes was considered referable. Main Outcome Measures Area under the receiver operating characteristic curve (AUC), sensitivity and specificity. Results In the internal validation dataset, the AUC, sensitivity and specificity of the DLA for neovascular AMD was 0.995, 96.7%, 96.4%, respectively. Testing against the independent external dataset achieved an AUC, sensitivity and specificity of 0.967, 100% and 93.4%, respectively. More than 60% of false positive cases displayed other macular pathologies. Amongst the false negative cases (internal validation dataset only), over half (57.2%) proved to be undetected detachment of the neurosensory retina or RPE layer. Conclusions and Relevance This DLA shows robust performance for the detection of neovascular AMD amongst retinal images from a multi-ethnic sample and under different imaging protoc

关键词： deep-learning algorithm age-related macular degeneration retinal-imaging

来源：评论

学校读者我要写书评

暂无评论

Method research on equilibrium cycle reloading pattern optimization for the HPR1000 reactor core using deep learning-enhanced multi-objective genetic algorithm

引用

ANNALS OF NUCLEAR ENERGY 2025年 220卷

作者： Butt, Muhammad Kamran Wan, Chenghui Wei, Luo Khan, Izat Cao, Liangzhi Xi An Jiao Tong Univ Sch Nucl Sci & Technol 28 Xianning West Rd Xian 710049 Shaanxi Peoples R China

A novel methodology for designing and optimizing a PWR's equilibrium cycle reloading pattern using deep learning and a multi-objective genetic algorithm (MOGA) has been developed. The deep-learning model efficiently and accurately predicted reactor physics parameters, particularly fuel assembly burnups at the end of the cycle (EOC), and formed a fitness function. The fitness function takes the absolute difference between the conformable fuel assemblies' burnups at the beginning of the cycle (BOC) and the EOC, which narrows down the potential equilibrium reloading patterns. The deep-learning model was coupled with MOGA, which simultaneously optimized multiple objectives for the design and optimization of an equilibrium cycle. Applied to the HPR1000 reactor, the method achieved the first equilibrium cycle length of 473.1 Effective Full Power Days (EFPDs) and an average of 471.1 EFPDs over ten cycles, meeting all the parameters of reactor safety design criteria.

关键词： deep-learning algorithm Multi-Objective Genetic algorithm Equilibrium-cycle optimization HPR1000

来源：评论

学校读者我要写书评

暂无评论

Research on the prediction method of fatigue stress response of the bogie end structure based on data-driven approaches

引用

ENGINEERING FAILURE ANALYSIS 2025年 175卷

作者： Song, Ye Liu, Chaotao Wu, Pingbo Wang, Xiangyang Dai, Huanyun Qi, Yayun Southwest Jiaotong Univ State Key Lab Rail Transit Vehicle Syst Chengdu 610031 Peoples R China Lenovo Syst Integrat Shenzhen Co Ltd Shenzheng 518108 Peoples R China Chongqing Jiaotong Univ Sch Mechanotron & Vehicle Engn Chongqing 400074 Peoples R China

As the mileage of railway vehicles continues to increase, the fatigue failure problem of the suspension components at the end of the bogie frame has become increasingly prominent. Clearly defining the fatigue failure mechanism of the suspension components at the end of the bogie frame and predicting their fatigue life have become crucial issues that urgently need to be addressed in vehicle maintenance and operation. When evaluating structural fatigue damage, the accuracy of stress measurement points has a decisive effect on the evaluation results of fatigue damage. However, ensuring the accuracy of structural stress measurement points and conducting long-term monitoring poses challenges. In this paper, the simulation technology is first used to identify the high-risk areas of structural fatigue damage. Subsequently, vibration line tests and stress tests are carried out on the relevant structures to verify the causes of structural failure. Then, a deep-learning algorithm is adopted to develop a method for detecting the structural stress of suspension components based on the acceleration data at the end of the bogie frame. This method is used to evaluate the fatigue damage during long-term operation and solve the problem that it is difficult to diagnose the faults of the suspension components at the end of the bogie frame. This method trains a deep-learning model with the historical data of vibration acceleration and stress, establishes the corresponding relationship between vibration acceleration and stress, and realizes the indirect detection of structural stress.

关键词： The suspension components Structural fatigue damage Stress and vibration test deep-learning algorithm Structural stress detection

来源：评论

学校读者我要写书评

暂无评论

Long-Range Disorder MetaSurface Enabled High-Performance One-Shot Ultraviolet Full-Stokes Polarimeter

引用

LASER & PHOTONICS REVIEWS 2025年第1期19卷

作者： Huang, Shanshan Xian, Shilin Peng, Jialong Yang, Xiu Du, Jinglei Hou, Yidong Sichuan Univ Chengdu 610065 Sichuan Peoples R China Sichuan Univ Sci & Engn Sch Phys & Elect Engn Zigong 643000 Peoples R China Natl Univ Def Technol Coll Adv Interdisciplinary Studies Changsha 410073 Peoples R China Natl Univ Def Technol Hunan Prov Key Lab Novel NanoOptoelect Informat Ma Changsha 410073 Peoples R China Natl Univ Def Technol Nanhu Laser Lab Changsha 410073 Peoples R China Univ Elect Sci & Technol China Sch Elect Sci & Engn Chengdu 610054 Peoples R China Sichuan Univ Key Lab High Energy Dens Phys & Technol Minist Educ Chengdu 610064 Peoples R China

The rapidly-developed nanophotonics enable the realization of highly-precise, ultra-compact full Stokes polarimeters. However, realizing large-area well-designed structures with complex chiral morphology and subwavelength size is still a challenge to the current micro/nano-engineering technology. Here, one high-performance, ultra-compact, and one-shot full-Stokes polarimeters in the ultraviolent waveband for the first time based on the long-range disorder chiral shells fabricated by the micro-sphere lithography are experimentally demonstrated. This chiral-shell monolayer owns strong and distinct optical chirality and anisotropy between the shells in adjacent micro-domains and thus leads to different photo-electric responses to the incident polarized lights for the photodetectors placed underneath. Through employing the residual convolutional neural network to extract the Stokes parameter S, a small detection averaged mean square error (MSE) of <0.5% from 316 nm to 410 nm is realized, and the minimum MSEs at 361 nm can reach recorded values of approximate to 0.02% (S-1), 0.017% (S-2), and 0.014% (S-3). The influence of exposure time and pixel number, and the system stability are systematically investigated. This work brings new inspiration for the disorder structures based on Bottom-Up methods, high-performance polarimeters, and polarization imaging devices.

关键词： chiral shells deep-learning algorithm full-stokes polarimeters microsphere lithography ultraviolent chiral metamaterials

来源：评论

学校读者我要写书评

暂无评论

Integrating the deep learning and multi-objective genetic algorithm to the reloading pattern optimization of HPR1000 reactor core

引用

NUCLEAR ENGINEERING AND DESIGN 2024年 428卷

作者： Butt, Muhammad Kamran Cao, Liangzhi Wan, Chenghui Lei, Kaihui Khan, Izat Xi An Jiao Tong Univ Sch Nucl Sci & Technol 28 Xianning West Rd Xian 710049 Shaanxi Peoples R China

The deep learning and multi-objective genetic algorithm were employed to optimize the fuel reloading pattern for HPR1000, a state-of-the-art nuclear power reactor designed and operated in China, also known as Hualong-1. In this study, the deep-learning algorithm was applied to establish the rapid evaluator for fuel-reloading patterns, for which the random samples between fuel-reloading patterns and corresponding key core parameters were generated by our home-developed nuclear-design code, named Bamboo-C. The advanced machine-learning platform TensorFlow was utilized for the deep-learning model. Then, the multi-objective genetic algorithm was applied to search the optimal fuel-reloading patterns, combined with the rapid evaluator to evaluate the key core parameters in a very short time. The DAKOTA toolkit was employed for optimization using a multi-objective genetic algorithm, for which the cycle length and power-peak factors were selected as the target parameters to establish the fitness function. For verification and application, the above method has been applied to the fuel- reloading optimization for Cycle 2 of HPR1000 operated in China. The optimization pattern results in an extension of the cycle length by about 21 EFPD (Effective Full Power Day), keeping all the key safety parameters satisfying corresponding safety criteria.

关键词： deep-learning algorithm Multi-objective genetic algorithm HPR1000 Bamboo-C

来源：评论

学校读者我要写书评

暂无评论

On flow regime transition in trickle bed: Development of a novel deep-learning-assisted image analysis method

引用

AICHE JOURNAL 2020年第2期66卷 e16833-e16833页

作者： Wang, Chao Li, Shaoshuo Yang, Yao Huang, Zhengliang Sun, Jingyuan Wang, Jingdai Yang, Yongrong Liao, Zuwei Jiang, Binbo Zhejiang Univ Coll Chem & Biol Engn Zhejiang Prov Key Lab Adv Chem Engn Manufacture T Hangzhou 310027 Zhejiang Peoples R China Zhejiang Univ State Key Lab Chem Engn Coll Chem & Biol Engn Hangzhou Zhejiang Peoples R China

An image analysis method was developed based on deep-learning algorithms to extract phase fractions quantitatively in a rectangular trickle bed, and the average identification error was lower than 5%. Furthermore, the flow regime transition in the trickle bed was studied. In trickle-to-pulse flow transition, the trickle flow could be further classified into the stable trickle flow and accelerated one. The SD of liquid fractions and the peak width at half-height of the probability density curve of liquid fractions were close to zero in stable trickle flow, increased rapidly in accelerated trickle flow, and remained approximately constant in pulse flow. In bubble-to-pulse flow transition, dispersed bubbles in bubble flow induced the outliers outside the upper boundary of the boxplot of gas fraction, while alternative appearance of gas-rich zone and liquid-rich zone in pulse flow induced outliers outside both the upper and lower boundaries of the boxplot of gas fraction.

关键词： deep-learning algorithm flow regime transition phase fraction quantitative image analysis method trickle bed

来源：评论

学校读者我要写书评

暂无评论

Mobile Sleep Stage Analysis Using Multichannel Wearable Devices Integrated with Stretchable Transparent Electrodes

引用

ACS Sensors 2025年 2025 May 15页

作者： Um, Hyun-Kyung Noh, Eunseo Yoo, Chaehwa Lee, Hyang Woon Kang, Je-Won Lee, Byoung Hoon Lee, Jung-Rok Department of Mechanical and Biomedical Engineering Ewha Womans University Seoul 03760 South Korea Graduate Program in Smart Factory Ewha Womans University Seoul 03760 South Korea Department of Chemical Engineering and Materials Science Ewha Womans University Seoul 03760 South Korea Graduate Program in System Health Science and Engineering Ewha Womans University Seoul 03760 South Korea School of Electrical Engineering Chungbuk National University Cheongju 28644 South Korea Department of Neurology and Medical Science School of Medicine Ewha Womans University Seoul 03760 South Korea Ewha Medical Research Institute Ewha Womans University Seoul 03760 South Korea Department of Electronic and Electrical Engineering Ewha Womans University Seoul 03760 South Korea

The prevalence of sleep disorders in the aging population and the importance of sleep quality for health have emphasized the need for accurate and accessible sleep monitoring solutions. Polysomnography (PSG) remains the clinical gold standard for diagnosing sleep disorders;however, its discomfort and inconvenience limit its accessibility. To address these issues, a wearable device (WD) integrated with stretchable transparent electrodes (STEs) is developed in this study for multisignal sleep monitoring and artificial intelligence (AI)-driven sleep staging. Utilizing conductive and flexible STEs, the WD records multiple biological signals (electroencephalogram [EEG], electrooculogram [EOG], electromyogram [EMG], photoplethysmography, and motion data) with high precision and low noise, comparable to PSG (<4 μV_RMS). It achieves a 73.2% accuracy and a macro F1 score of 0.72 in sleep staging using an AI model trained on multisignal inputs. Notably, accuracy marginally improves when using only the EEG, EOG, and EMG channels, which may simplify future device designs. This WD offers a compact, multisignal solution for at-home sleep monitoring, with the potential for use as an evaluation tool for personalized sleep therapies. © 2025 American Chemical Society.

关键词： biopotential signals deep-learning algorithm electroencephalography sleep stage analysis stretchable transparent electrodes wearable devices

来源：评论

学校读者我要写书评

暂无评论

Predictive Cruise Cloud Control Scheme Design onNotable Vehicles-Under the Perspective ofCyber-Physical Systems

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2024年第7期25卷 6796-6810页

作者： Lin, Jingdong Li, Yang Xiao, Hongzhao Chongqing Univ Sch Automat Chongqing 400044 Peoples R China Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore

The notable vehicles (NV) predictive cruise cloud control (PCCC) system has great potential in improving the performances of driving safety and energy saving by virtue of advantages on the decent calculation and optimal control capability of cloud control system (CCS). The traditional PCC systems lack the interoperable integration control of cyber layer and physical plane, limited by the range of perception and on-board computing ability, so that the information acquisition and processing are quite restrained, which impede the improvement of control performances to a certain extent. In this study, we propose a CCS layered architecture based on cyber-physical systems (CPS), which effectively addresses the problems of real-time multi-sources heterogeneous information utilization by PCC systems and insufficient computational planning capability of solver of NV. Then, based on the digital information in the cloud control basic platform, we propose an intensified PCCC framework in the hierarchical architecture application platform. In this framework, we introduce a modified deep-learning traffic prediction model to predict the traffic state after pre-factorization processing and the optimization algorithm solver to optimize the proposed scheme under the restrictive constraints via using the prediction information and digital information. We fully evaluate the effectiveness of the architecture under various traffic scenarios through simulation. The results show that the proposed layered architecture has implementation feasibility and great potential in the application of vehicle-cloud layered control.

关键词： Computer architecture Transportation Safety Predictive models Optimization Cloud computing Real-time systems Cyber-physical systems predictive cruise cloud control cloud control system traffic prediction deep-learning algorithm vehicle-cloud layered control

来源：评论

学校读者我要写书评

暂无评论

Hydraulic Fracturing Fluid Interactions With Steel, Cement, and Shale Measured

引用

JPT, Journal of Petroleum Technology 2022年第9期74卷 91-93页

作者： Xiong, Wei McAdams, Brandon Stuckman, Mengling National Energy Technology Laboratory United States

_. This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 206537, “Development of deep Transformer-Based Models for Long-Term Prediction of Transient Production of Oil Wells,” by Ildar R. Abdrakhmanov, Evgenii A. Kanin, SPE, and Sergei A. Boronin, Skolkovo Institute of Science and Technology, et al. The paper has not been peer reviewed._. The authors of the complete paper propose a novel approach to data-driven modeling of transient production of oil wells, applying transformer-based neural networks trained on multivariate time series composed of various parameters of oil wells measured during exploitation. By tuning machine-learning models for a single well (ignoring the effect of neighboring wells) on open-source field data sets, the authors of the paper demonstrate that the transformer-based method outperforms recurrent neural networks (RNNs) with long- and short-term memory (LSTM) and gated recurrent-unit cells in the forecasting of bottomhole pressure dynamics. Introduction. The authors apply a novel deep-learning algorithm called a transformer to build surrogate models for simulations of well performance. Transformer architecture initially was developed for natural-language processing problems. However, in recent years, researchers have adapted transformers for time-series forecasting. In contrast with RNNs, the transformer does not process data sequentially. Instead, it handles the entire sequence by a multihead self-attention mechanism. The transformer is much more computationally efficient than an RNN because its training can be performed using graphics processing *** networks allow for application of the transfer-learning technique. First, the model is trained to predict the target parameter (e.g., bottomhole pressure) using the production data from a certain well. Next, the tuned weights are used to initialize the training procedure on the data from a target well. In other words, the fine-tuning is p

关键词： reservoir simulation reservoir surveillance upstream oil & gas transformer production control transient production neural network production monitoring transformer-based model aid prediction deep-learning algorithm Artificial intelligence Formation Evaluation & Management Information Management and Systems Neural networks Reservoir Simulation Well & Reservoir Surveillance and Monitoring

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：