检索结果-内蒙古大学图书馆

Enhancing MR Image Segmentation with Realistic Adversarial data Augmentation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Chen, Chen Qin, Chen Ouyang, Cheng Li, Zeju Wang, Shuo Qiu, Huaqi Chen, Liang Tarroni, Giacomo Bai, Wenjia Rueckert, Daniel Department of Computing Imperial College London United Kingdom Institute for Digital Communications School of Engineering University of Edinburgh United Kingdom Digital Medicine Research Centre School of Basic Medical Sciences Fudan University China Shanghai Key Laboratory of MICCAI Shanghai China CitAI Research Centre Department of Computer Science City University of London United Kingdom Department of Brain Sciences Imperial College London United Kingdom Data Science Institute Imperial College London United Kingdom Klinikum Rechts der Isar Technical University of Munich Germany

The success of neural networks on medical image segmentation tasks typically relies on large labeled datasets for model training. However, acquiring and manually labeling a large medical image set is resource-intensive, expensive, and sometimes impractical due to data sharing and privacy issues. To address this challenge, we propose AdvChain, a generic adversarial data augmentation framework, aiming at improving both the diversity and effectiveness of training data for medical image segmentation tasks. AdvChain augments data with dynamic data augmentation, generating randomly chained photo-metric and geometric transformations to resemble realistic yet challenging imaging variations to expand training data. By jointly optimizing the data augmentation model and a segmentation network during training, challenging examples are generated to enhance network generalizability for the downstream task. The proposed adversarial data augmentation does not rely on generative networks and can be used as a plug-in module in general segmentation networks. It is computationally efficient and applicable for both low-shot supervised and semi-supervised learning. We analyze and evaluate the method on two MR image segmentation tasks: cardiac segmentation and prostate segmentation with limited labeled data. Results show that the proposed approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications. Copyright © 2021, The Authors. All rights reserved.

关键词： Image segmentation

3D deep learning for enhanced atom probe tomography analysis of nanoscale microstructures

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yu, Jiwei Wang, Zhangwei Saksena, Aparna Wei, Shaolou Wei, Ye Colnaghi, Timoteo Marek, Andreas Rampp, Markus Song, Min Gault, Baptiste Li, Yue State Key Laboratory of Powder Metallurgy Central South University Changsha410083 China Max-Planck-Institut für Eisenforschung GmbH Max-Planck-Straße 1 Düsseldorf40237 Germany Ecole Polytechnique Fédérale de Lausanne School of Engineering Rte Cantonale Lausanne1015 Switzerland Max Planck Computing and Data Facility Gießenbachstraße 2 Garching85748 Germany Department of Materials Imperial College South Kensington LondonSW7 2AZ United Kingdom

Quantitative analysis of microstructural features on the nanoscale, including precipitates, local chemical orderings (LCOs) or structural defects (e.g. stacking faults) plays a pivotal role in understanding the mechanical and physical responses of engineering materials. Atom probe tomography (APT), known for its exceptional combination of chemical sensitivity and sub-nanometer resolution, primarily identifies microstructures through compositional segregations. However, this fails when there is no significant segregation, as can be the case for LCOs and stacking faults. Here, we introduce a 3D deep learning approach, AtomNet, designed to process APT point cloud data at the single-atom level for nanoscale microstructure extraction, simultaneously considering compositional and structural information. AtomNet is showcased in segmenting L12-type nanoprecipitates from the matrix in an AlLiMg alloy, irrespective of crystallographic orientations, which outperforms previous methods. AtomNet also allows for 3D imaging of L10-type LCOs in an AuCu alloy, a challenging task for conventional analysis due to their small size and subtle compositional differences. Finally, we demonstrate the use of AtomNet for revealing 2D stacking faults in a Co-based superalloy, without any defected training data, expanding the capabilities of APT for automated exploration of hidden microstructures. AtomNet pushes the boundaries of APT analysis, and holds promise in establishing precise quantitative microstructure-property relationships across a diverse range of metallic materials. © 2024, CC BY.

关键词： Stacking faults

Inequality-Type Zhang Equivalency Originating from Neural Dynamics for Time-Varying Problems Solving

学校读者我要写书评

暂无评论

Inequality-Type Zhang Equivalency Originating from Neural Dy...

IEEE International Conference on Mechatronics and Automation

作者： Yunong Zhang Liangjie Ming Min Yang Binbin Qiu Zhonghua Li School of Data and Computer Science Sun Yat-sen University Guangzhou P. R. China Research Institute of Sun Yat-sen University in Shenzhen Shenzhen P. R. China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou P. R. China School of Electronics and Information Technology Sun Yat-sen University Guangzhou P. R. China School of Intelligent Systems Engineering Sun Yat-sen University Guangzhou P. R. China

ISBN: (数字)9781728164168

ISBN: (纸本)9781728164175

In this paper, a concept of Zhang equivalency (ZE) is presented, and its type of inequality to inequality (simply termed, inequality type) is further discussed. By applying Zhang neural dynamics method, we thus propose the order-1, order2, and order-3 ZE general formulas of such an inequality type, respectively. Then, the inequalities in four different situations are investigated, i.e., the robot situation, matrix situation (including matrix-vector situation), state-control situation, and state situation. We propose the corresponding order-1, order-2, and order-3 ZE formulas to prepare for solving corresponding time-varying inequality-related problems.

关键词： Mechatronics Automation Conferences Linear matrix inequalities Problem-solving Robots

FAST Observations of Four Comets to Search for the Molecular Line Emissions between 1.0 and 1.5 GHz Frequencies

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Long-Fei Tsai, Chao-Wei Li, Jian-Yang Yang, Bin Li, Di Duan, Yan Hsia, Chih-Hao Pan, Zhichen Qian, Lei Quan, Donghui Jiang, Xue-Jian Li, Xiaohu Zhao, Ruining Zuo, Pei School of Physics and Electronic Science Guizhou Normal University Guiyang550025 China Guizhou Provincial Key Laboratory of Radio Astronomy and Data Processing Guiyang550025 China Research Center for Astronomical Computing Zhejiang Laboratory Hangzhou311100 China National Astronomical Observatories Chinese Academy of Sciences Beijing100101 China Institute for Frontiers in Astronomy and Astrophysics Beijing Normal University Beijing102206 China Key Laboratory of Radio Astronomy and Technology Chinese Academy of Sciences A20 Datun Road Chaoyang District Beijing100101 China School of Atmospheric Sciences Sun Yat-sen University Guangdong Zhuhai China Núcleo de Astronomía Facultad de Ingenieríay Ciencias Universidad Diego Portales Chile Department of Astronomy College of Physics and Electronic Engineering Qilu Normal University 2 Wenbo Road Zhangqiu District Jinan250200 China Department of Electronic and Optical Engineering Space Engineering University Beijing China Laboratory for Space Research Faculty of Science The University of Hong Kong Hong Kong Xinjiang Astronomical Observatory Chinese Academy of Sciences No. 150 Science 1-Street Urumqi830011 China CAS Key Laboratory of Optical Astronomy National Astronomical Observatories Chinese Academy of Sciences Beijing100101 China School of Astronomy and Space Sciences University of Chinese Academy of Sciences Beijing100049 China

We used the Five-hundred-meter Aperture Spherical radio Telescope (FAST) to search for the molecular emissions in the L-band between 1.0 and 1.5 GHz toward four comets, C/2020 F3 (NEOWISE), C/2020 R4 (ATLAS), C/2021 A1 (Leonard), and 67P/Churyumov-Gerasimenko during or after their perihelion passages. Thousands of molecular transition lines fall in this low-frequency range, many attributed to complex organic or prebiotic molecules. We conducted a blind search for the possible molecular lines in this frequency range in those comets and could not identify clear signals of molecular emissions in the data. Although several molecules have been detected at high frequencies of great than 100 GHz in comets, our results confirm that it is challenging to detect molecular transitions in the L-band frequency ranges. The non-detection of L-band molecular lines in the cometary environment could rule out the possibility of unusually strong lines, which could be caused by the masers or non-LTE effects. Although the line strengths are predicted to be weak, for FAST, using the ultra-wide bandwidth receiver and improving the radio frequency interference environments would enhance the detectability of those molecular transitions at low frequencies in the future. Copyright © 2024, The Authors. All rights reserved.

关键词： Molecules

Large AI Models in Health Informatics: Applications, Challenges, and the Future

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Qiu, Jianing Li, Lin Sun, Jiankai Peng, Jiachuan Shi, Peilun Zhang, Ruiyang Dong, Yinzhao Lam, Kyle Lo, Frank P.-W. Xiao, Bo Yuan, Wu Wang, Ningli Xu, Dong Lo, Benny The Department of Computing Imperial College London United Kingdom The Department of Biomedical Engineering The Chinese University of Hong Kong Hong Kong Ltd. Hong Kong The Department of Informatics King’s College London United Kingdom The School of Engineering Stanford University United States The Department of Engineering Science University of Oxford United Kingdom The Faculty of Engineering The University of Hong Kong Hong Kong The Department of Surgery and Cancer Imperial College London United Kingdom The Hamlyn Centre for Robotic Surgery Imperial College London United Kingdom Beijing Tongren Eye Center Beijing Tongren Hospital Capital Medical University China Beijing Ophthalmology & Visual Sciences Key Laboratory Beijing China The Department of Electrical Engineering and Computer Science United States The Christopher S. Bond Life Sciences Center University of Missouri United States The Facualty of Medicine Imperial College London United Kingdom

Large AI models, or foundation models, are models recently emerging with massive scales both parameter-wise and data-wise, the magnitudes of which can reach beyond billions. Once pretrained, large AI models demonstrate impressive performance in various downstream tasks. A prime example is ChatGPT, whose capability has compelled people’s imagination about the far-reaching influence that large AI models can have and their potential to transform different domains of our lives. In health informatics, the advent of large AI models has brought new paradigms for the design of methodologies. The scale of multi-modal data in the biomedical and health domain has been ever-expanding especially since the community embraced the era of deep learning, which provides the ground to develop, validate, and advance large AI models for breakthroughs in health-related areas. This article presents a comprehensive review of large AI models, from background to their applications. We identify seven key sectors in which large AI models are applicable and might have substantial influence, including 1) bioinformatics;2) medical diagnosis;3) medical imaging;4) medical © 2023, CC BY-NC-ND.

关键词： Medical imaging

Exploiting Gabor Feature Extraction Method for Chinese Character Writing Quality Evaluation

学校读者我要写书评

暂无评论

Journal of Physics: Conference Series 2020年第1期1575卷

作者： Zhixiao Wang Wenyao Yan Mingtao Guo Jiulong Zhang School of Computer Science and Engineering Xi'an University of Technology Xi'an 710048 China Shaanxi Key Laboratory Xi'an of Network Computing and Security Technology China School of Data Science and Computer Xi'an innovation college of Yan'an University Xi'an 710100 China

The automatic evaluation of Chinese character writing quality has a wide application prospect. Most of the existing evaluation methods of Chinese character writing quality are based on radical segmentation and feature judgment, which require the high accuracy of Chinese character segmentation. However, there are many problems in the real handwriting, such as continuous writing, uneven strength of writing, personalized writing style and so on, which lead to the difficulty of segmentation in the ordinary handwriting. To solve the above problems, we propose an effective method based on image texture where the uniformity of writing lines and writing style is taken as an effective criterion. In our method, Gabor transform is used to extract the image features of writing samples, and finally the statistical learning method of support vector machine is used to effectively evaluate the writing quality. Experiments on multiple real datasets including CHAED show that our method is effective and accurate. The advantage of this method is that it does not need to segment fonts, and the cost of global feature extraction is small.

关键词：

Circuit area optimization of multi-output MPRM based on ERWOA algorithm

学校读者我要写书评

暂无评论

Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics 2023年第5期49卷 1193-1200页

作者： He, Juncai He, Zhenxue Wang, Fushun Huo, Zhisheng Xiao, Limin Intelligent Agricultural Equipment Research Institute Hebei Agricultural University Baoding071001 China Key Laboratory of Agricultural Big Data of Hebei Province Hebei Agricultural University Baoding071001 China High Performance Computing Platform Beihang University Beijing100191 China School of Software Beihang University Beijing100191 China School of Computer Science and Engineering Beihang University Beijing100191 China

Mixed polarity Reed-Muller (MPRM) circuit area optimization has become a research hotspot in the field of integrated circuit design. It is a combinatorial optimization, aiming at finding the MPRM expression with the least number of terms among many MPRM expressions. A explosion strategy and restart strategy based whale optimization algorithm (ERWOA) is proposed. In addition, a multi-output MPRM circuit area optimization method is proposed, which uses the improved whale algorithm and the improved polarity conversion algorithm to search for the MPRM circuit with the least number of AND terms. Results on the MCNC Benchmark circuits show that the proposed algorithm increases the conversion efficiency by 99.93% and 99.96% at most, compared with the mixed polarity and inter-polarity conversion algorithms based on the list technology, respectively. Compared with the genetic algorithm and the artificial bee colony algorithm, the improved whale optimization algorithm saves the circuit area up to 18.32% with an average of 5.54%, and 14.41% with an average of 5.00%, respectively. © 2023 Beijing University of Aeronautics and Astronautics (BUAA). All rights reserved.

关键词： Genetic algorithms

Coexistence of insulator-like paramagnon and metallic spin-orbit exciton modes in SrIrO3

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Paris, Eugenio Zhang, W. Tseng, Y. Efimenko, A. Sahle, C. Strocov, V.N. Skoropata, E. Rolfs, K. Shang, T. Lyu, J. Pomjakushina, E. Medarde, M. Rønnow, H.M. Normand, B. Radovic, M. Schmitt, T. PSI Center for Photon Sciences Villigen-PSI CH-5232 Switzerland European Synchrotron Radiation Facility 71 Avenue des Martyrs Grenoble38043 France PSI Center for Neutron and Muon Sciences Villigen-PSI CH-5232 Switzerland School of Physics and Electronic Science East China Normal University Shanghai200241 China CAS Key Laboratory of Magnetic Materials and Devices Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences Ningbo315201 China LausanneCH-1015 Switzerland PSI Center for Scientific Computing Theory and Data Villigen-PSI CH-5232 Switzerland

We probe the spectrum of elementary excitations in SrIrO3 by using heterostructured [(SrIrO3)m/(SrTiO3)l] samples to approach the bulk limit. Our resonant inelastic x-ray scattering (RIXS) measurements at the Ir L3-edge reveal a robust low-lying collective magnetic mode with an antiferromagnetic (AF) dispersion similar to the insulators Sr2IrO4 and Sr3Ir2O7, albeit with a large gap and much larger linewidth. At higher energies we find the spin-orbit exciton, also strongly broadened, but with an inverted dispersion and doubled periodicity that are controlled by the charge hopping. These results demonstrate that the AF paramagnon persists, somewhat counterintuitively, far into the metallic regime of the insulator-metal transition driven by the degree of confinement in the heterostructure. We conclude that these two excitations, which are contrasting but coexisting hallmarks of strong AF pseudospin and charge fluctuations in a spin-orbit-coupled Mott-Slater material, are properties intrinsic to the ground state of semimetallic perovskite SrIrO3. © 2024, CC BY.

关键词： Perovskite

Uncertainty-inspired Open Set Learning for Retinal Anomaly Identification

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Meng Lin, Tian Wang, Lianyu Lin, Aidi Zou, Ke Xu, Xinxing Zhou, Yi Peng, Yuanyuan Meng, Qingquan Qian, Yiming Deng, Guoyao Wu, Zhiqun Chen, Junhong Lin, Jianhong Zhang, Mingzhi Zhu, Weifang Zhang, Changqing Zhang, Daoqiang Goh, Rick Siow Mong Liu, Yong Pang, Chi Pui Chen, Xinjian Chen, Haoyu Fu, Huazhu 1 Fusionopolis Way #16-16 Connexis Singapore138632 Singapore Joint Shantou International Eye Center Shantou University the Chinese University of Hong Kong Medical College Shantou University Guangdong515041 China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Jiangsu211100 China National Key Laboratory of Fundamental Science on Synthetic Vision the College of Computer Science Sichuan University Sichuan610065 China School of Electronics and Information Engineering Soochow University Jiangsu215006 China School of Biomedical Engineering Anhui Medical University Anhui230032 China Longchuan People’s Hospital Heyuan China Puning People’s Hospital Jieyang China Haifeng PengPai Memory Hospital Shanwei China College of Intelligence and Computing Tianjin University Tianjin300350 China Department of Ophthalmology and Visual Sciences The Chinese University of Hong Kong Hong Kong State Key Laboratory of Radiation Medicine and Protection Soochow University Suzhou215006 China

Failure to recognize samples from the classes unseen during training is a major limitation of artificial intelligence in the real-world implementation for recognition and classification of retinal anomalies. We established an uncertainty-inspired open-set (UIOS) model, which was trained with fundus images of 9 retinal conditions. Besides assessing the probability of each category, UIOS also calculated an uncertainty score to express its confidence. Our UIOS model with thresholding strategy achieved an F1 score of 99.55%, 97.01% and 91.91% for the internal testing set, external target categories (TC)-JSIEC dataset and TC-unseen testing set, respectively, compared to the F1 score of 92.20%, 80.69% and 64.74% by the standard AI model. Furthermore, UIOS correctly predicted high uncertainty scores, which would prompt the need for a manual check in the datasets of non-target categories retinal diseases, low-quality fundus images, and non-fundus images. UIOS provides a robust method for real-world screening of retinal anomalies. © 2023, CC BY.

关键词： Uncertainty analysis