检索结果-内蒙古大学图书馆

A SEMI-SUPERVISED LEARNING APPROACH FOR B-LINE DETECTION IN LUNG ULTRASOUND IMAGES

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Tianqi Anantrasirichai, Nantheera Karakuş, Oktay Allinovi, Marco Achim, Alin Visual Information Lab University of Bristol Bristol United Kingdom School of Computer Science and Informatics Cardiff University Cardiff United Kingdom Nephrology Dialysis and Transplantation Careggi University Hospital Florence Italy

Studies have proved that the number of B-lines in lung ultrasound images has a strong statistical link to the amount of extravascular lung water, which is significant for hemodialysis treatment. Manual inspection of B-lines requires experts and is time-consuming, whilst designing automatic methods is currently problematic because of the lack of ground truth. Therefore, in this paper, we propose a novel semi-supervised learning method for the B-line detection task based on contrastive learning. Through multi-level unsupervised learning on unlabelled lung ultrasound images, the features of the artefacts are learnt. In the downstream task, we introduce a fine-tuning process on a small number of labelled images using the EIoU-based loss function. Apart from reducing the data labelling workload, the proposed method shows a superior performance compared to model-based methods with the recall of 91.43%, the accuracy of 84.21% and the F1 score of 91.43%. Copyright © 2022, The Authors. All rights reserved.

关键词： Unsupervised learning

Towards Quantum Graph Neural Networks: An Ego-Graph Learning Approach

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ai, Xing Sun, Luzhe Yan, Junchi Zhang, Zhihong Hancock, Edwin R. School of Informatics Xiamen University Fujian Xiamen China Computer Science Department in University of Chicago ChicagoIL United States Department of Computer Science and Engineering MOE Key Lab of Artificial Intelligence Shanghai Jiao Tong University Shanghai200240 China University of York York United Kingdom

Quantum machine learning is a fast-emerging field that aims to tackle machine learning using quantum algorithms and quantum computing. Due to the lack of physical qubits and an effective means to map real-world data from Euclidean space to Hilbert space, most of these methods focus on quantum analogies or process simulations rather than devising concrete architectures based on qubits. In this paper, we propose a novel hybrid quantum-classical algorithm for graph-structured data, which we refer to as the Ego-graph based Quantum Graph Neural Network (egoQGNN). egoQGNN implements the GNN theoretical framework using the tensor product and unity matrix representation, which greatly reduces the number of model parameters required. When controlled by a classical computer, egoQGNN can accommodate arbitrarily sized graphs by processing ego-graphs from the input graph using a modestly-sized quantum device. The architecture is based on a novel mapping from real-world data to Hilbert space. This mapping maintains the distance relations present in the data and reduces information loss. Experimental results show that the proposed method outperforms competitive state-of-the-art models with only 1.68% parameters compared to those models. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

Reversible Data Hiding Using Convolutional Neural Network and Digital Signal Processing Techniques

学校读者我要写书评

暂无评论

Reversible Data Hiding Using Convolutional Neural Network an...

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Peng Qingmei Li Shaozi Lin Yaojin Yu Xiehua Key Laboratory of Big Data and Artificial Intelligence University of Fujian Province MinNan Science and Technology University Quanzhou China School of Informatics Xiamen University Xiamen China School of Computer Science Lab of Data Science and Intelligence Application Minnan Normal University Zhangzhou China

Reversible data hiding is a data hiding process in which data is encrypted into an image as a means of authentication, and hidden data is used to hide the original image, and RDH is often used in areas where data authenticity and data preservation are important, such as legal/criminal investigations, medical images. In this study, this paper designed a novel reversible data hiding system that utilizes the Steghide algorithm to encrypt images and hide data, as well as an artificial neural network to recover embedded text and structure. Based on the analysis of the test results, it was concluded that the RDH scheme proposed in this article has the same or higher system accuracy and reliability as the existing RDH scheme. To further refine the system results in future studies, this paper proposes to increase the number of training iterations and the number of images in the training dataset to further improve the accuracy and reliability of the development.

关键词： Training System testing Signal processing algorithms Digital signal processing Reliability Cryptography Time factors

Training Hacks and a Frugal Man's Net with Application to Glioblastoma Segmentation 6

学校读者我要写书评

暂无评论

Training Hacks and a Frugal Man's Net with Application to Gl...

6th International Conference on Advanced Technologies for Signal and Image Processing, ATSIP 2022

作者： Abdallah, Jawher Ben Marrakchi-Kacem, Linda Rekik, Islem Université de la Manouba Institut Supérieur de Biotechnologie de Sidi Thabet Tunis Tunisia Université Tunis El Manar L3S Ecole Nationale d'Ingénieurs de Tunis Tunis Tunisia Istanbul Technical University Basira Lab Faculty of Computer and Informatics Istanbul Turkey University of Dundee School of Science and Engineering Computing United Kingdom

ISBN: (纸本)9781665451161

In this paper, we investigate the effectiveness of training a sparse Neural Network on a limited number of samples in the context of brain tumor segmentation. Nowadays, Deep Learning architectures are getting deeper, more sophisticated and environmentally unfriendly in an effort to improve their segmentation performance. We use a brain tumor segmentation dataset and apply simple practices to reduce the needed computational resources to allow cheap and fast training. We also present a lighter, cheaper version of the U-Net dubbed Frugal U-Net stemming from our investigation on how far we can push the original U-Net by decreasing its parameter count using Depth-Wise Separable Convolutions instead of regular ones, all the while preserving the minimum levels of accuracy required in Medical Imaging. Our methodology is useful in clinical facilities where high-computation resources are limited. © 2022 IEEE.

关键词： Medical imaging

Automatic Polyp Segmentation with Multiple Kernel Dilated Convolution Network

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tomar, Nikhil Kumar Srivastava, Abhishek Bagci, Ulas Jha, Debesh School of Informatics and Computer Science Indira Gandhi National Open University India Computer Vision and Pattern Recognition Unit Indian Statistical Institute India Machine and Hybrid Intelligence Lab Department of Radiology Feinberg School of Medicine Northwestern University United States

The detection and removal of precancerous polyps through colonoscopy is the primary technique for the prevention of colorectal cancer worldwide. However, the miss rate of colorectal polyp varies significantly among the endoscopists. It is well known that a computer-aided diagnosis (CAD) system can assist endoscopists in detecting colon polyps and minimize the variation among endoscopists. In this study, we introduce a novel deep learning architecture, named MKDCNet, for automatic polyp segmentation robust to significant changes in polyp data distribution. MKDCNet is simply an encoder-decoder neural network that uses the pre-trained ResNet50 as the encoder and novel multiple kernel dilated convolution (MKDC) block that expands the field of view to learn more robust and heterogeneous representation. Extensive experiments on four publicly available polyp datasets and cell nuclei dataset show that the proposed MKDCNet outperforms the state-of-the-art methods when trained and tested on the same dataset as well when tested on unseen polyp datasets from different distributions. With rich results, we demonstrated the robustness of the proposed architecture. From an efficiency perspective, our algorithm can process at (≈ 45) frames per second on RTX 3090 GPU. MKDCNet can be a strong benchmark for building real-time systems for clinical colonoscopies. The code of the proposed MKDCNet is available at https://***/nikhilroxtomar/MKDCNet. © 2022, CC BY.

关键词： Convolution

Instruction Multi-Constraint Molecular Generation Using a Teacher-Student Large Language Model

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhou, Peng Wang, Jianmin Li, Chunyan Wang, Zixu Liu, Yiping Sun, Siqi Lin, Jianxin Wei, Leyi Cai, Xibao Lai, Houtim Liu, Wei Wang, Longyue Liu, Yuansheng Zeng, Xiangxiang College of Information Science and Engineering Hunan University Changsha China The Interdisciplinary Graduate Program in Integrative Biotechnology Yonsei University Incheon21983 Korea Republic of School of Informatics Yunnan Normal University Kunming650500 China Department of Computer Science University of Tsukuba Tsukuba3058577 Japan Research Institute of Intelligent Complex Systems Fudan University Shanghai200433 China Shanghai AI Laboratory Shanghai200232 China Centre for Artificial Intelligence driven Drug Discovery Faculty of Applied Science Macao Polytechnic University China School of Informatics Xiamen University Xiamen China Tencent AI Lab China

While various models and computational tools have been proposed for structure and property analysis of molecules, generating molecules that conform to all desired structures and properties remains a challenge. Here, we introduce a multi-constraint molecular generation large language model, TSMMG, which, akin to a student, incorporates knowledge from various small models and tools, namely, the 'teachers'. To train TSMMG, we construct a large set of text-molecule pairs by extracting molecular knowledge from these 'teachers', enabling it to generate novel molecules that conform to the descriptions through various text prompts. We experimentally show that TSMMG remarkably performs in generating molecules meeting complex, natural language-described property requirements across two-, three-, and four-constraint tasks, with an average molecular validity of over 99% and success ratio of 82.58%, 68.03%, and 67.48%, respectively. The model also exhibits adaptability through zero-shot testing, creating molecules that satisfy combinations of properties that have not been encountered. It can comprehend text inputs with various language styles, extending beyond the confines of outlined prompts, as confirmed through empirical validation. Additionally, the knowledge distillation feature of TSMMG contributes to the continuous enhancement of small models, while the innovative approach to dataset construction effectively addresses the issues of data scarcity and quality, which positions TSMMG as a promising tool in the domains of drug discovery and materials science. Copyright © 2024, The Authors. All rights reserved.

关键词： Students

A Novel Delegated Proof of Work Consensus Protocol

学校读者我要写书评

暂无评论

A Novel Delegated Proof of Work Consensus Protocol

2021 International Conference on Artificial Intelligence for Cyber Security Systems and Privacy, AI-CSP 2021

作者： Kara, Mostefa Laouid, Abdelkader Bounceur, Ahcene Lalem, Farid Alshaikh, Muath Kebache, Romaissa Sayah, Zaoui El Oued University LIAP Laboratory El Oued39000 Algeria University of Western Brittany UBO Lab-STICC UMR CNRS Brest6285 France Saudi Electronic University College of Computing and Informatics Computer Science Department Riyadh11673 Saudi Arabia

ISBN: (纸本)9781665467148

Blockchain is a new technology for storing and transmitting information evolving without centralized control, it uses peer-to-peer networks and allows the constitution of replicated and distributed registers, these registers are secured thanks to cryptography by the linking of blocks to each other. As part of solving the challenges that exist to develop this technology, we present this article introducing significant improvements to this efficient mechanism. These improvements consist of proposing a consensus algorithm called Delegated Proof of Work (DPoW). Thanks to this algorithm, we offer a significant reduction in the very large amount of energy currently consumed in the mining process, we reduce the domination of an element with great computing power, we reinforce security against the 51 % attack, reducing its percentage to 67 %, and we also reduce the influence of false identities created within the network (Sybil Attack). On the other hand, we propose a consensus model for permissioned blockchain, this model is based on the encryption of random values and offers total security to reach a consensus. © 2021 IEEE.

关键词： Blockchain

Skeletonization Quality Evaluation: Geometric Metrics for Point Cloud Analysis in Robotics

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wen, Qingmeng Lai, Yu-Kun Ji, Ze Tafrishi, Seyed Amir School of Engineering Cardiff University CF24 3AA United Kingdom School of Computer Science and Informatics Cardiff University CF24 4AG United Kingdom Lab School of Engineering Cardiff University Queen’s Buildings The Parade CardiffCF24 3AA United Kingdom

Skeletonization is a powerful tool for shape analysis, rooted in the inherent instinct to understand an object’s morphology. It has found applications across various domains, including robotics. Although skeletonization algorithms have been studied in recent years, their performance is rarely quantified with detailed numerical evaluations. This work focuses on defining and quantifying geometric properties to systematically score the skeletonization results of point cloud shapes across multiple aspects, including topological similarity, boundedness, centeredness, and smoothness. We introduce these representative metric definitions along with a numerical scoring framework to analyze skeletonization outcomes concerning point cloud data for different scenarios, from object manipulation to mobile robot navigation. Additionally, we provide an open-source tool to enable the research community to evaluate and refine their skeleton models. Finally, we assess the performance and sensitivity of the proposed geometric evaluation methods from various robotic applications. Copyright © 2025, The Authors. All rights reserved.

关键词： Geometry

Non-volatile optical switch based on Sb2Se3-loaded three-mode interference

学校读者我要写书评

暂无评论

Optics letters 2025年第12期50卷 4022-4025页

作者： Xinxin Zhao Xiaolong Liu Yujiao Tao Lei Yang Kai Xu Hongtao Lin Peipeng Xu Wei Zhang Faculty of Electrical Engineering and Computer Science Ningbo University Ningbo 315211 China Key Laboratory of Photoelectric Detecting Materials and Devices of Zhejiang Province Ningbo 315211 China The State Key Lab of Brain-Machine Intelligence Key Laboratory of Micro-Nano Electronics and Smart System of Zhejiang Province College of Information Science and Electronic Engineering Zhejiang University Hangzhou 310027 China State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information Technology Chinese Academy of Sciences Shanghai 200050 China

In this Letter, we propose a non-volatile switch based on multimode interference in an SbSe-loaded waveguide. By implementing low-loss phase control within a multimode interference coupler, this design effectively manages the self-imaging phase shift length during a single phase change cycle. The fabricated multiplexed optical switch exhibits impressive performance characteristics, including a low insertion loss of ∼ 1.5 dB, a compact length of ∼35 μm, and minimal cross talk below -6 dB over a 35 nm bandwidth for both the amorphous and crystalline states. The ability to reconfigure light routing via SbSe phase transitions opens new avenues for optical field-programmable gate arrays and high-density photonic processors.

关键词： Crosstalk Effective refractive index Field programmable gate arrays Multimode interference Optical neural systems Phase shift