检索结果-内蒙古大学图书馆

WaveFormer: A Wavelet Transformer for Parkinson Disease’s Retinal Layer Segmentation in OCT

学校读者我要写书评

暂无评论

WaveFormer: A Wavelet Transformer for Parkinson Disease’s R...

International Joint Conference on Neural Networks (IJCNN)

作者： Yanlin Chen Xiaoqing Zhang Tianao Wang Chen Tang Haili Ye Jiang Liu Department of Computer Science and Engineering Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China Shenzhen Institute of Advanced Technology Center for High Performance Computing and Shenzhen Key Laboratory of Intelligent Bioinformatics Chinese Academy of Sciences Shenzhen China School of Ophthalmology and Optometry Wenzhou Medical University Wenzhou China The Oujiang Laboratory The Affiliated Eye Hospital Wenzhou Medical University Wenzhou China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Pathology symptoms of Parkinson disease (PD) are different from those of retinal diseases in the retinal layers, which are subtle. However, segmenting pathology information of PD from retinal layers automatically based on optical coherence tomography (OCT) images has not been studied before. Although existing Transformer-based segmentation methods have achieved good segmentation results, they have limitations in capturing local context information. Convolutional neural networks (CNNs) can construct local context dependencies among pixels, which is complementary to Transformers. Particularly, edge information extraction is significant for accurate retinal layer segmentation, which is ignored by both Transformers and CNNs but can be captured by frequency domain learning methods. To fully leverage the advantages of Transformers, CNNs, and frequency domain learning methods, we propose a Wavelet Transformer (WaveFormer) for retinal layer segmentation based on OCT images. In the WaveFormer, we design a Wavelet Spatial Attention block to exploit the potential of frequency information. Based on these advantages, WaveFormer can be data-efficient in limited OCT images of PD. The experimental results on the OCT-PD segmentation dataset show that our WaveFormer outperforms existing Transformers and CNNs. For example, WaveFormer outperforms Swin-UNet by 3.41% of IoU.

关键词： Learning systems Image segmentation Pathology Wavelet domain Image edge detection Frequency-domain analysis Transformers

Towards Convergence of Blockchain and Self-sovereign Identity for Privacy-Preserving Secure Federated Learning 1

学校读者我要写书评

暂无评论

3rd International Conference on Big Data and Security, ICBDS 2021

作者： Haque, Rakib Ul Hasan, A. S. M. Touhidul Daria, Apubra Qu, Qiang Jiang, Qingshan School of Computer Science and Technology University of Chinese Academy of Sciences Shijingshan District Beijing100049 China Department of Computer Science and Engineering University of Asia Pacific Dhaka1205 Bangladesh Institute of Automation Research and Engineering Dhaka1205 Bangladesh Shenzhen Key Laboratory for High Performance Data Mining Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China

ISBN: (数字)9789811908521

ISBN: (纸本)9789811908514

More and more researchers are eager to train their Machine Learning (ML) model by the distributed dataset in a federated manner. However, there are numerous privacy and security concerns in federated learning, i.e., adversarial attacks, authentication, and model inversion attacks. In this paper, we integrate self-sovereign identity for privacy-preserving secure federated learning. The proposed framework represents identity management and authentication scheme for edge devices, which are participated in FL. It guarantees participants’ privacy, where users’ private data are not shared with the centralized training module. The trust triangle and Blockchain ensure the authenticity of the distributed federated learning, where data privacy is handled with Differential Privacy. Result analysis is done based on measured score and time complexity. In terms of accuracy, BCWD, HDD, and DD achieve 94.69 %, 81.05 % and 78.00 %, respectively on standard ML, whereas 94.45 %, 80.59 % and 76.43 % on ϵ DP-based FL with privacy budget, ϵ = 3. For the time complexity measurement, the total number of transactions employed in the system is 1000. Transactions time and their verification took 3323.15 s. Moreover, the proposed system shows robust performance in both cases. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Blockchain

Fourier domain adaptation for nighttime pedestrian detection using Faster R-CNN

学校读者我要写书评

暂无评论

Fourier domain adaptation for nighttime pedestrian detection...

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Peggy Joy Lu Jen-Hui Chuang Department of Computer Science National Center for High-Performance Computing (NCHC) National Yang Ming Chiao Tung University Hsinchu Taiwan Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan

ISBN: (数字)9781665484855

ISBN: (纸本)9781665484862

An efficient domain adaptation scheme is presented in this paper for nighttime pedestrian detection using Faster R-CNN. First, we adopt Fourier domain adaptation on training data by replacing low-frequency spectrum of source data (RGB images) with that of target data (infrared images). Such approach is more efficient compared with existing state-of-the-art methods of domain adaptation for object detection, as it does not require adversarial learning, or adding extra components to the Faster R-CNN. In addition, a simple preprocessing of intensity scaling is empirically selected among several image enhancement algorithms for testing data. Experimental results demonstrate that performance improvements of up to 30% and 10% can be achieved with the above processes for training data and testing data, respectively, for an indoor scene with poor illumination condition (while other processes may actually lower the performance).

关键词： Circuits and systems Data preprocessing Training data Lighting Transforms Object detection Adversarial machine learning

Potrika: Raw and Balanced Newspaper Datasets in the Bangla Language with Eight Topics and Five Attributes

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ahmad, Istiak AlQurashi, Fahad Mehmood, Rashid Department of Computer Science Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia High Performance Computing Center King Abdulaziz University Jeddah21589 Saudi Arabia

Knowledge is central to human and scientific developments. Natural Language Processing (NLP) allows automated analysis and creation of knowledge. Data is a crucial NLP and machine learning ingredient. The scarcity of open datasets is a well-known problem in machine and deep learning research. This is very much the case for textual NLP datasets in English and other major world languages. For the Bangla language, the situation is even more challenging and the number of large datasets for NLP research is practically nil. We hereby present Potrika, a large single-label Bangla news article textual dataset curated for NLP research from six popular online news portals in Bangladesh (Jugantor, Jaijaidin, Ittefaq, Kaler Kontho, Inqilab, and Somoyer Alo) for the period 2014-2020. The articles are classified into eight distinct categories (National, Sports, International, Entertainment, Economy, Education, Politics, and science & Technology) providing five attributes (News Article, Category, Headline, Publication Date, and Newspaper Source). The raw dataset contains 185.51 million words and 12.57 million sentences contained in 664,880 news articles. Moreover, using NLP augmentation techniques, we create from the raw (unbalanced) dataset another (balanced) dataset comprising 320,000 news articles with 40,000 articles in each of the eight news categories. Potrika contains both the datasets (raw and balanced) to suit a wide range of NLP research. By far, to the best of our knowledge, Potrika is the largest and the most extensive dataset for news classification. © 2022, CC BY.

关键词： Deep learning

Machine and Deep Learning Methods with Manual and Automatic Labelling for News Classification in Bangla Language

学校读者我要写书评

暂无评论

arXiv 2022年

Research in Natural Language Processing (NLP) has increasingly become important due to applications such as text classification, text mining, sentiment analysis, POS tagging, named entity recognition, textual entailment, and many others. This paper introduces several machine and deep learning methods with manual and automatic labelling for news classification in the Bangla language. We implemented several machine (ML) and deep learning (DL) algorithms. The ML algorithms are Logistic Regression (LR), Stochastic Gradient Descent (SGD), Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbour (KNN), used with Bag of Words (BoW), Term Frequency-Inverse Document Frequency (TF-IDF), and Doc2Vec embedding models. The DL algorithms are Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and Convolutional Neural Network (CNN), used with Word2vec, Glove, and FastText word embedding models. We develop automatic labelling methods using Latent Dirichlet Allocation (LDA) and investigate the performance of single-label and multi-label article classification methods. To investigate performance, we developed from scratch Potrika, the largest and the most extensive dataset for news classification in the Bangla language, comprising 185.51 million words and 12.57 million sentences contained in 664, 880 news articles in eight distinct categories, curated from six popular online news portals in Bangladesh for the period 2014-2020. GRU and Fasttext with 91.83% achieve the highest accuracy for manually-labelled data. For the automatic labelling case, KNN and Doc2Vec at 57.72% and 75% achieve the highest accuracy for single-label and multi-label data, respectively. The methods developed in this paper are expected to advance research in Bangla and other languages. © 2022, CC BY.

关键词： Classification (of information)

Dual-scale Enhanced and Cross-generative Consistency Learning for Semi-supervised Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Gu, Yunqi Zhou, Tao Zhang, Yizhe Zhou, Yi He, Kelei Gong, Chen Fu, Huazhu PCA Laboratory The School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China The School of Computer Science and Engineering Southeast University Nanjing211189 China The Medical School Nanjing University Nanjing210023 China The National Institute of Healthcare Data Science Nanjing University Nanjing210023 China The Institute of High Performance Computing A*STAR Singapore

Medical image segmentation plays a crucial role in computer-aided diagnosis. However, existing methods heavily rely on fully supervised training, which requires a large amount of labeled data with time-consuming pixel-wise annotations. Moreover, accurately segmenting lesions poses challenges due to variations in shape, size, and location. To address these issues, we propose a novel Dual-scale Enhanced and Cross-generative consistency learning framework for semi-supervised medical image Segmentation (DEC-Seg). First, we propose a Cross-level Feature Aggregation (CFA) module that integrates cross-level adjacent layers to enhance the feature representation ability across different resolutions. To address scale variation, we present a scale-enhanced consistency constraint, which ensures consistency in the segmentation maps generated from the same input image at different scales. This constraint helps handle variations in lesion sizes and improves the robustness of the model. Furthermore, we propose a cross-generative consistency scheme, in which the original and perturbed images can be reconstructed using cross-segmentation maps. This consistency constraint allows us to mine effective feature representations and boost the segmentation performance. To further exploit the scale information, we propose a Dual-scale Complementary Fusion (DCF) module that integrates features from two scale-specific decoders operating at different scales to help produce more accurate segmentation maps. Extensive experimental results on multiple medical segmentation tasks (polyp, skin lesion, and brain glioma) demonstrate the effectiveness of our DECSeg against other state-of-the-art semi-supervised segmentation approaches. The implementation code will be released at https://***/taozh2017/DECSeg. Copyright © 2023, The Authors. All rights reserved.

关键词： Self-supervised learning

Enhancing and Adapting in the Clinic: Source-free Unsupervised Domain Adaptation for Medical Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Heng Lin, Ziqin Qiu, Zhongxi Li, Zinan Niu, Ke Guo, Na Fu, Huazhu Hu, Yan Liu, Jiang The Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Shenzhen China The Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China The Institute of High Performance Computing Agency for Science Technology and Research Singapore Computer School Beijing Information Science and Technology University Beijing China The School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

Medical imaging provides many valuable clues involving anatomical structure and pathological characteristics. However, image degradation is a common issue in clinical practice, which can adversely impact the observation and diagnosis by physicians and algorithms. Although extensive enhancement models have been developed, these models require a well pre-training before deployment, while failing to take advantage of the potential value of inference data after deployment. In this paper, we raise an algorithm for source-free unsupervised domain adaptive medical image enhancement (SAME), which adapts and optimizes enhancement models using test data in the inference phase. A structure-preserving enhancement network is first constructed to learn a robust source model from synthesized training data. Then a teacher-student model is initialized with the source model and conducts source-free unsupervised domain adaptation (SFUDA) by knowledge distillation with the test data. Additionally, a pseudo-label picker is developed to boost the knowledge distillation of enhancement tasks. Experiments were implemented on ten datasets from three medical image modalities to validate the advantage of the proposed algorithm, and setting analysis and ablation studies were also carried out to interpret the effectiveness of SAME. The remarkable enhancement performance and benefits for downstream tasks demonstrate the potential and generalizability of SAME. The code is available at https://***/liamheng/Annotation-free-Medical-Image-Enhancement. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

Characterizing the spatial potential of a surface electrode ion trap

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Qin, Qingqing Chen, Ting Zhang, Xinfang Ou, Baoquan Zhang, Jie Wu, Chunwang Xie, Yi Wu, Wei Chen, Pingxing College of Science National University of Defense Technology Changsha410073 China Hunan Key Laboratory of Mechanism and Technology of Quantum Information Hunan Changsha410073 China Institute for Quantum Information State Key Laboratory of High Performance Computing College of Computer Science National University of Defense Technology Changsha410073 China

The accurate characterization of the spatial potential generated by a planar electrode in a surface-type Paul trap is of great interest. To achieve this, we employ a simple yet highly precise parametric expression to describe the spatial field of a rectangular-shaped electrode. Based on this, an optimization method is introduced to precisely characterize the axial electric field intensity created by the powered electrode and the stray field. In contrast to existing methods, various types of experimental data, such as the equilibrium position of ions in a linear string, equilibrium positions of single trapped ions and trap frequencies, are utilized for potential estimation in order to mitigate systematic errors. This approach offers significant flexibility in voltage settings for data collection, making it particularly well-suited for surface electrode traps where ion probe trapping height may vary with casual voltage settings. In our demonstration, we successfully minimized the discrepancy between experimental observations and model predictions to an impressive extent. The relative errors of secular frequencies were suppressed within ±0.5%, and the positional error of ions was limited to less than 1.2 µm, all surpassing those achieved by existing methodologies. Copyright © 2023, The Authors. All rights reserved.

关键词： Electrodes

Uncertainty-informed Mutual Learning for Joint Medical Image Classification and Segmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ren, Kai Zou, Ke Liu, Xianjie Chen, Yidi Yuan, Xuedong Shen, Xiaojing Wang, Meng Fu, Huazhu National Key Laboratory of Fundamental Science on Synthetic Vision Sichuan University Sichuan China College of Computer Science Sichuan University Sichuan China Department of Radiology West China Hospital Sichuan University Sichuan China College of Mathematics Sichuan University Sichuan China Institute of High Performance Computing Agency for Science Technology and Research Singapore

Classification and segmentation are crucial in medical image analysis as they enable accurate diagnosis and disease monitoring. However, current methods often prioritize the mutual learning features and shared model parameters, while neglecting the reliability of features and performances. In this paper, we propose a novel Uncertainty-informed Mutual Learning (UML) framework for reliable and interpretable medical image analysis. Our UML introduces reliability to joint classification and segmentation tasks, leveraging mutual learning with uncertainty to improve performance. To achieve this, we first use evidential deep learning to provide image-level and pixel-wise confidences. Then, an uncertainty navigator is constructed for better using mutual features and generating segmentation results. Besides, an uncertainty instructor is proposed to screen reliable masks for classification. Overall, UML could produce confidence estimation in features and performance for each link (classification and segmentation). The experiments on the public datasets demonstrate that our UML outperforms existing methods in terms of both accuracy and robustness. Our UML has the potential to explore the development of more reliable and explainable medical image analysis models. © 2023, CC BY.

关键词： Uncertainty analysis