检索结果-内蒙古大学图书馆

Breast Cancer Immunohistochemical Image Generation: a Benchmark dataset and Challenge Review

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhu, Chuang Liu, Shengjie Yu, Zekuan Xu, Feng Aggarwal, Arpit Corredor, Germán Madabhushi, Anant Qu, Qixun Fan, Hongwei Li, Fangda Li, Yueheng Guan, Xianchao Zhang, Yongbing Singh, Vivek Kumar Akram, Farhan Sarker, Md Mostafa Kamal Shi, Zhongyue Jin, Mulan The School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China Academy for Engineering and Technology Fudan University Shanghai China Beijing Chaoyang Hospital Capital Medical University Beijing China Biomedical Engineering Department Georgia Tech and Emory University United States The Atlanta Veterans Affairs Medical Center United States Biomedical Engineering Department Emory University United States Data Science Institute Imperial College London United Kingdom The School of Computer and Electrical Engineering Purdue University United States The School of Computer Science and Technology Harbin Institute of Technology Shenzhen China The Department of Computer Engineering and Mathematics Rovira I Virgili University Spain The Department of Pathology and Clinical Bioinformatics Erasmus Medical Center Netherlands The Institute of Biomedical Engineering University of Oxford Oxford United Kingdom

For invasive breast cancer, immunohistochemical (IHC) techniques are often used to detect the expression level of human epidermal growth factor receptor-2 (HER2) in breast tissue to formulate a precise treatment plan. From the perspective of saving manpower, material and time costs, directly generating IHC-stained images from Hematoxylin and Eosin (H&E) stained images is a valuable research direction. Therefore, we held the breast cancer immunohistochemical image generation challenge, aiming to explore novel ideas of deep learning technology in pathological image generation and promote research in this field. The challenge provided registered H&E and IHC-stained image pairs, and participants were required to use these images to train a model that can directly generate IHC-stained images from corresponding H&E-stained images. We selected and reviewed the five highest-ranking methods based on their PSNR and SSIM metrics, while also providing overviews of the corresponding pipelines and implementations. In this paper, we further analyze the current limitations in the field of breast cancer immunohistochemical image generation and forecast the future development of this field. We hope that the released dataset and the challenge will inspire more scholars to jointly study higher-quality IHC-stained image generation. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

Neural Brain: A Neuroscience-inspired Framework for Embodied Agents

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Liu, Jian Shi, Xiongtao Nguyen, Thai Duy Zhang, Haitian Zhang, Tianxiang Sun, Wei Li, Yanjie Vasilakos, Athanasios V. Iacca, Giovanni Khan, Arshad Ali Kumar, Arvind Cho, Jae Won Mian, Ajmal Xie, Lihua Cambria, Erik Wang, Lin School of Electrical and Electronic Engineering Nanyang Technological University Singapore School of Robotics Hunan University China School of Intelligence Science and Engineering Harbin Institute of Technology Shenzhen China Department of Information and Communication Technology University of Agder Norway Department of Information Engineering and Computer Science University of Trento Italy Elm Company London United Kingdom Division of Computational Science and Technology KTH Royal Institute of Technology Sweden School of Artificial Intelligence and Data Science Sejong University Korea Republic of Department of Computer Science the University of Western Australia Australia College of Computing and Data Science Nanyang Technological University Singapore

The rapid evolution of artificial intelligence (AI) has shifted from static, data-driven models to dynamic systems capable of perceiving and interacting with real-world environments. Despite advancements in pattern recognition and symbolic reasoning, current AI systems, such as large language models, remain disembodied, unable to physically engage with the world. This limitation has driven the rise of embodied AI, where autonomous agents, such as humanoid robots, must navigate and manipulate unstructured environments with human-like adaptability. At the core of this challenge lies the concept of Neural Brain, a central intelligence system designed to drive embodied agents with human-like adaptability. A Neural Brain must seamlessly integrate multimodal sensing and perception with cognitive capabilities. Achieving this also requires an adaptive memory system and energy-efficient hardware-software co-design, enabling real-time action in dynamic environments. This paper introduces a unified framework for the Neural Brain of embodied agents, addressing two fundamental challenges: (1) defining the core components of Neural Brain and (2) bridging the gap between static AI models and the dynamic adaptability required for real-world deployment. To this end, we propose a biologically inspired architecture that integrates multimodal active sensing, perception-cognition-action function, neuroplasticity-based memory storage and updating, and neuromorphic hardware/software optimization. Furthermore, we also review the latest research on embodied agents across these four aspects and analyze the gap between current AI systems and human intelligence. By synthesizing insights from neuroscience, we outline a roadmap towards the development of generalizable, autonomous agents capable of human-level intelligence in real-world scenarios. Our project page is at Neural-Brain-for-Embodied-Agents. Copyright © 2025, The Authors. All rights reserved.

关键词： Anthropomorphic robots

Deep Learning Approaches for Enhancing Content Retrieval

学校读者我要写书评

暂无评论

Deep Learning Approaches for Enhancing Content Retrieval

Smart Generation Computing, Communication and Networking (SMART GENCON), International Conference on

作者： Saniya Khurana Shweta Singh M K Ranganathaswamy G. Renuka Mandar P. Diwakar Sandeep Kumar Jain Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and technology Chitkara University Punjab India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Department of Mechanical Engineering Faculty of Engineering and Technology JAIN (Deemed-to-be University) Ramnagar Karnataka India Department of Electronics and Communication Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai Department of Artificial Intelligence & Data Science Vishwakarma Institute of Information Technology Pune India Department of Electrical Engineering Vivekananda Global University Jaipur India

Deep learning strategies for content material retrieval have been explored and updated to cope with the want for accurate search effects. This machine up to date is used to date apprehend and examine text-up-to-date content material from sources, up-to-date webpages, or virtual up-todate motion pictures. As the quantity and complexity of online content will increase, so up-to-date have the want for higher up-to-date up to date those resources. Recent advancements in deep gaining knowledge have enabled better content retrieval structures that are up to date, updated correctly, and speedy to understand statistics from one-of-a-kind sources. Such structures include natural language processing (NLP), sentiment analysis, entity reputation, and up-to-date updated reputation. Those deep studying technologies have been used to decorate content material retrieval by enhancing search effects' accuracy, robotically updated extracting applicable statistics, categorizing information, and identifying patterns. Additionally, deep-getting, updated systems are up-to-date and quick to examine large volumes of data, generating and relevant consequences than up-to-date methods. In summary, deep up-to-date techniques can beautify the retrieval system of content material efficiently, enhancing the accuracy and pace of seek effects.

关键词：

Secuyolo: Secure Object Detection Framework With Darknet-53 and CBAM in Edge Devices

学校读者我要写书评

暂无评论

Concurrency and Computation: Practice and Experience 2025年第15-17期37卷

作者： P. Lakshmiramana A. Vadivel Computer Science & Engineering GITAM School of Technology (Research Scholar) & Madanapalle Institute of Technology & Science (Assistant Professor) GITAM (Deemed to Be University) Dodballapura Karnataka India Madanapalle Institute of Technology & Science Angallu India Artificial Intelligence and Data Science GITAM School of Technology GITAM (Deemed to Be University) Dodballapura Karnataka India

Object detection in edge computing has emerged as a pivotal solution for processing sensor data locally, thereby minimizing latency and bandwidth consumption. However, this introduces significant privacy concerns, as edge servers may exploit sensitive sensor data. Traditional frameworks often overlook data security and privacy protection, presenting challenges in balancing privacy with computational efficiency in resource-constrained environments. To address these issues and improve performance, we propose SecuYOLO CBAMNet53, a novel framework combining Secure YOLO V3 with Darknet-53 and the CBAM. This model enhances feature extraction capabilities, ensuring privacy protection and accurate real-time object detection on edge devices. Darknet-53 serves as a deep feature extractor, while CBAM boosts discriminative power through channel-wise and spatial attention, improving detection accuracy and handling complex scenes. It integrates YOLOv3 with security measures like data encryption, secure deployment, and differential privacy, safeguarding sensitive information during inference. By classifying data as public or private for smart IoT applications, SecuYOLO CBAMNet53 optimizes performance and prevents privacy breaches. The proposed SecuYOLO, employing Darknet-53 with CBAM as its backbone, achieves outstanding performance in both an average recall of 67.62 and an average precision of 51.2. The model's dynamic recalibration of feature maps and robust security mechanisms ensure reliable object detection and privacy protection, making it suitable for real-world applications.

关键词： attention module data privacy deep learning edge computing object detection YOLOv3

Broad learning systems: An overview of recent advances, applications, challenges and future directions

学校读者我要写书评

暂无评论

Neurocomputing 2025年 641卷

作者： Chu, Yonghe Guo, Yanlong Ding, Weiping Cao, Heling Ping, Peng School of Artificial Intelligence and Computer Science Nantong University Jiangsu Nantong226019 China College of Information Science and Engineering Henan University of Technology Zhengzhou450001 China Faculty of Data Science City University of Macau 999078 China School of Transportation and Civil Engineering Nantong University Jiangsu Nantong226001 China

Broad learning system (BLS) as a novel training framework derived from the random vector functional link neural network (RVFLNN). Unlike RVFLNN, which directly applies raw data to network learning, BLS first transforms input data into feature nodes through feature mapping. These feature nodes are then nonlinearly transformed into enhanced nodes. Both feature nodes and enhanced nodes are concatenated and connected to the output layer, and the corresponding output weights are derived via pseudo-inverse. Due to its shallow architecture and the need to train only the output weights, BLS achieves highly efficient learning ***, when adding new nodes, BLS does not require retraining from scratch;it only adjusts the weights associated with the new nodes. Different from deep learning, BLS aims to expand the broad rather than the depth of the neural network to solve complex problems. It not only overcomes the drawback of the long-training process in deep learning but also enables the rapid incremental construction of network models. Therefore, BLS has received extensive attention from both the academic and industrial communities. In view of this, this paper conducts a systematic review of BLS. First, we outline the research background of the BLS. Then, we elaborate on the relevant concepts and definitions of BLS. Furthermore, recent advances in BLS are introduced. In addition, we present the extensive applications of BLS in various fields. Finally, several possible development directions for BLS are proposed. © 2025 Elsevier B.V.

关键词： Optical character recognition

Phase-Change Metasurfaces for Dynamic Image Display and Information Encryption

学校读者我要写书评

暂无评论

Physical Review Applied 2022年第4期18卷 044078-044078页

作者： Tingting Liu Zhou Han Junyi Duan Shuyuan Xiao Institute of Photonic Chips University of Shanghai for Science and Technology Shanghai 200093 People’s Republic of China Centre for Artificial-Intelligence Nanophotonics School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai 200093 People’s Republic of China Institute for Advanced Study Nanchang University Nanchang 330031 People’s Republic of China Jiangxi Key Laboratory for Microscale Interdisciplinary Study Nanchang University Nanchang 330031 People’s Republic of China

Optical metasurfaces enable us to engineer the electromagnetic space and control light propagation at an unprecedented level, offering a powerful tool to achieve modulation of light over multiple physical dimensions. Here, we demonstrate a Sb2S3 phase-change metasurface platform that allows active manipulation of both the amplitude and the phase. In particular, we implement dynamic nanoprinting and holographic image display through the tuning of the crystallization levels of this phase-change material. The Sb2S3 nanobricks tailored to perform the amplitude, geometric, and propagation-phase modulation constitute the dynamic meta-atoms in the multiplexed metasurfaces. Using the incident polarizations as decoding keys, the encoded information can be reproduced in the form of a nanoprinted grayscale image in the near field and two holographic images in the far field. These images can be switched on and off by taking advantage of the reversible tunability of the Sb2S3 nanostructure between the amorphous and crystalline states. The proposed phase-change metasurfaces featuring manifold information and multifold encryption promise ultracompact data storage with high capacity and high security, which suggests an exciting direction for modern cryptography and security applications.

关键词： Imaging & optical processing Nanophotonics Phase-change materials Chalcogenides Metasurfaces Finite-difference time-domain method Optical computing Optical techniques

ANA: Ant nesting algorithm for optimizing real-world problems

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Rashid, Deeam Najmadeen Hama Rashid, Tarik A. Mirjalili, Seyedali Department of Computer Science and Engineering School of Science and Engineering University of Kurdistan Hewler KRG Erbil44001 Iraq Centre for Artificial Intelligence Research and Optimisation Torrens University Australia Yonsei Frontier Lab Yonsei University Seoul Korea Republic of

In this paper, a novel swarm intelligent algorithm is proposed called ant nesting algorithm (ANA). The algorithm is inspired by Leptothorax ants and mimics the behavior of ants searching for positions to deposit grains while building a new nest. Although the algorithm is inspired by the swarming behavior of ants, it does not have any algorithmic similarity with the ant colony optimization (ACO) algorithm. It is worth mentioning that ANA is considered a continuous algorithm that updates the search agent position by adding the rate of change (e.g., step or velocity). ANA computes the rate of change differently as it uses previous, current solutions, fitness values during the optimization process to generate weights by utilizing the Pythagorean theorem. These weights drive the search agents during the exploration and exploitation phases. The ANA algorithm is benchmarked on 26 well-known test functions, and the results are verified by a comparative study with genetic algorithm (GA), particle swarm optimization (PSO), dragonfly algorithm (DA), five modified versions of PSO, whale optimization algorithm (WOA), salp swarm algorithm (SSA), and fitness dependent optimizer (FDO). ANA outperformances these prominent metaheuristic algorithms on several test cases and provides quite competitive results. Finally, the algorithm is employed for optimizing two well-known real-world engineering problems: antenna array design and frequency-modulated synthesis. The results on the engineering case studies demonstrate the proposed algorithm's capability in optimizing real-world problems. © 2021, CC BY.

关键词： Frequency modulation

Cooperative Sensing and Heterogeneous Information Fusion in VCPS: A Multi-agent Deep Reinforcement Learning Approach

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Xincao Liu, Kai Dai, Penglin Xie, Ruitao Cao, Jingjing Luo, Jiangtao The College of Computer Science Chongqing University Chongqing400040 China The School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu611756 China The National Engineering Laboratory of Integrated Transportation Big Data Application Technology Chengdu611756 China The College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China The School of Transportation and Logistics Engineering Wuhan University of Technology Hubei430063 China The Electronic Information and Networking Research Institute Chongqing University of Posts and Telecommunications Chongqing400065 China

Cooperative sensing and heterogeneous information fusion are critical to realize vehicular cyber-physical systems (VCPSs). This paper makes the first attempt to quantitatively measure the quality of VCPS by designing a new metric called Age of View (AoV). Specifically, we first present the system architecture where heterogeneous information can be cooperatively sensed and uploaded via vehicle-to-infrastructure (V2I) communications in vehicular edge computing (VEC). Logical views are constructed by fusing the heterogeneous information at edge nodes. Further, we formulate the problem by deriving a cooperative sensing model based on the multi-class M/G/1 priority queue, and defining the AoV by modeling the timeliness, completeness and consistency of the logical views. On this basis, a multi-agent deep reinforcement learning solution is proposed. In particular, the system state includes vehicle sensed information, edge cached information and view requirements. The vehicle action space consists of the sensing frequencies and uploading priorities of information. A difference-reward-based credit assignment is designed to divide the system reward, which is defined as the VCPS quality, into the difference reward for vehicles. Edge node allocates V2I bandwidth to vehicles based on predicted vehicle trajectories and view requirements. Finally, we build the simulation model and give a comprehensive performance evaluation, which conclusively demonstrates the superiority of the proposed solution. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

Inferring gene regulatory networks by hypergraph generative model

学校读者我要写书评

暂无评论

Cell Reports Methods 2025年第4期5卷 101026页

作者： Su, Guangxin Wang, Hanchen Zhang, Ying Wilkins, Marc R. Canete, Pablo F. Yu, Di Yang, Yang Zhang, Wenjie School of Computer Science and Engineering The University of New South Wales Sydney NSW Australia ARC Centre of Excellence for the Mathematical Analysis of Cellular Systems (MACSYS) Melbourne VIC Australia Australian Artificial Intelligence Institute The University of Technology Sydney Sydney NSW Australia School of Computer Science and Technology Zhejiang Gongshang University Zhejiang China Systems Biology Initiative School of Biotechnology and Biomolecular Sciences The University of New South Wales Sydney NSW Australia Frazer Institute Faculty of Health Medicine and Behaviour Sciences The University of Queensland Brisbane QLD Australia Ian Frazer Centre for Children's Immunotherapy Research Child Health Research Centre Faculty of Health Medicine and Behaviour Sciences The University of Queensland Brisbane QLD Australia

We present hypergraph variational autoencoder (HyperG-VAE), a Bayesian deep generative model that leverages hypergraph representation to model single-cell RNA sequencing (scRNA-seq) data. The model features a cell encoder with a structural equation model to account for cellular heterogeneity and construct gene regulatory networks (GRNs) alongside a gene encoder using hypergraph self-attention to identify gene modules. The synergistic optimization of encoders via a decoder improves GRN inference, single-cell clustering, and data visualization, as validated by benchmarks. HyperG-VAE effectively uncovers gene regulation patterns and demonstrates robustness in downstream analyses, as shown in B cell development data from bone marrow. Gene set enrichment analysis of overlapping genes in predicted GRNs confirms the gene encoder's role in refining GRN inference. Offering an efficient solution for scRNA-seq analysis and GRN construction, HyperG-VAE also holds the potential for extending GRN modeling to temporal and multimodal single-cell omics. © 2025 The Authors

关键词： CP: Systems biology gene regulatory networks hypergraph representation learning scRNA-seq