检索结果-内蒙古大学图书馆

Segmentation of Head and Neck Tumors Using Dual PET/CT Imaging:Comparative Analysis of 2D,2.5D,and 3D Approaches Using UNet Transformer

引用

computer Modeling in Engineering & sciences 2024年第12期141卷 2351-2373页

作者： Mohammed A.Mahdi Shahanawaj Ahamad Sawsan A.Saad Alaa Dafhalla Alawi Alqushaibi Rizwan Qureshi Information and Computer Science Department College of Computer Science and EngineeringUniversity of Ha’ilHa’il55476Saudi Arabia Software Engineering Department College of Computer Science and EngineeringUniversity of Ha’ilHa’il55476Saudi Arabia Computer Engineering Department College of Computer Science and EngineeringUniversity of Ha’ilHa’il55476Saudi Arabia Department of Computer and Information Sciences Universiti Teknologi PetronasSeri Iskandar32610Malaysia Center for Research in Computer Vision(CRCV) University of Central FloridaOrlandoFL 32816USA

The segmentation of head and neck(H&N)tumors in dual Positron Emission Tomography/Computed Tomogra-phy(PET/CT)imaging is a critical task in medical imaging,providing essential information for diagnosis,treatment planning,and outcome *** by the need for more accurate and robust segmentation methods,this study addresses key research gaps in the application of deep learning techniques to multimodal medical ***,it investigates the limitations of existing 2D and 3D models in capturing complex tumor structures and proposes an innovative 2.5D UNet Transformer model as a *** primary research questions guiding this study are:(1)How can the integration of convolutional neural networks(CNNs)and transformer networks enhance segmentation accuracy in dual PET/CT imaging?(2)What are the comparative advantages of 2D,2.5D,and 3D model configurations in this context?To answer these questions,we aimed to develop and evaluate advanced deep-learning models that leverage the strengths of both CNNs and *** proposed methodology involved a comprehensive preprocessing pipeline,including normalization,contrast enhancement,and resampling,followed by segmentation using 2D,2.5D,and 3D UNet Transformer *** models were trained and tested on three diverse datasets:HeckTor2022,AutoPET2023,and *** was assessed using metrics such as Dice Similarity Coefficient,Jaccard Index,Average Surface Distance(ASD),and Relative Absolute Volume Difference(RAVD).The findings demonstrate that the 2.5D UNet Transformer model consistently outperformed the 2D and 3D models across most metrics,achieving the highest Dice and Jaccard values,indicating superior segmentation *** instance,on the HeckTor2022 dataset,the 2.5D model achieved a Dice score of 81.777 and a Jaccard index of 0.705,surpassing other model *** 3D model showed strong boundary delineation performance but exhibited variability across datasets,while the

关键词： PET/CT imaging tumor segmentation weighted fusion transformer multi-modal imaging deep learning neural networks clinical oncology

来源：评论

学校读者我要写书评

暂无评论

LEARNING SEMANTIC PROXIES FROM VISUAL PROMPTS FOR PARAMETER-EFFICIENT FINE-TUNING IN DEEP METRIC LEARNING 12

LEARNING SEMANTIC PROXIES FROM VISUAL PROMPTS FOR PARAMETER-...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Ren, Li Chen, Chen Wang, Liqiang Hua, Kien Department of Computer Science University of Central Florida United States Center for Research in Computer Vision University of Central Florida United States

Deep Metric Learning (DML) has long attracted the attention of the machine learning community as a key objective. Existing solutions concentrate on fine-tuning the pre-trained models on conventional image datasets. As a result of the success of recent pre-trained models trained from larger-scale datasets, it is challenging to adapt the model to the DML tasks in the local data domain while retaining the previously gained knowledge. In this paper, we investigate parameter-efficient methods for fine-tuning the pre-trained model for DML tasks. In particular, we propose a novel and effective framework based on learning Visual Prompts (VPT) in the pre-trained vision Transformers (ViT). Based on the conventional proxy-based DML paradigm, we augment the proxy by incorporating the semantic information from the input image and the ViT, in which we optimize the visual prompts for each class. We demonstrate that our new approximations with semantic information are superior to representative capabilities, thereby improving metric learning performance. We conduct extensive experiments to demonstrate that our proposed framework is effective and efficient by evaluating popular DML benchmarks. In particular, we demonstrate that our fine-tuning method achieves comparable or even better performance than recent state-of-the-art full fine-tuning works of DML while tuning only a small percentage of total parameters. Code is available at https://***/Noahsark/ParameterEfficient-DML. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Exploring Parameter-Efficient Fine-Tuning to Enable Foundation Models in Federated Learning

Exploring Parameter-Efficient Fine-Tuning to Enable Foundati...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Sun, Guangyu Khalid, Umar Mendieta, Matias Wang, Pu Chen, Chen University of Central Florida Center for Research in Computer Vision OrlandoFL United States University of North Carolina at Charlotte Department of Computer Science CharlotteNC United States

ISBN: (纸本)9798350362480

Federated learning (FL) has emerged as a promising paradigm for enabling the collaborative training of models without centralized access to the raw data on local devices. In the typical FL paradigm (e.g., FedAvg), model weights are sent to and from the server each round to participating clients. Recently, the use of small pre-trained models has been shown to be effective in federated learning optimization and improving convergence. However, recent state-of-the-art pre-trained models are getting more capable but also have more parameters, known as the "Foundation Models."In conventional FL, sharing the enormous model weights can quickly put a massive communication burden on the system, especially if more capable models are employed. Can we find a solution to enable those strong and readily available pre-trained models in FL to achieve excellent performance while simultaneously reducing the communication burden? To this end, we investigate the use of parameter-efficient fine-tuning in federated learning and thus introduce a new framework: FedPEFT. Specifically, we systemically evaluate the performance of FedPEFT across a variety of client stability, data distribution, and differential privacy settings. By only locally tuning and globally sharing a small portion of the model weights, significant reductions in the total communication overhead can be achieved while maintaining competitive or even better performance in a wide range of federated learning scenarios, providing insight into a new paradigm for practical and effective federated systems. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A multi-feature-based intelligent redundancy elimination scheme for cloud-assisted health systems

引用

CAAI Transactions on Intelligence Technology 2024年第2期9卷 491-510页

作者： Ling Xiao Beiji Zou Xiaoyan Kui Chengzhang Zhu Wensheng Zhang Xuebing Yang Bob Zhang School of Computer Science and Engineering Central South UniversityChangshaChina Hunan Engineering Research Center of Machine Vision and Intelligent Medicine Central South UniversityChangshaChina The College of Literature and Journalism Central South UniversityChangshaChina Institute of Automation Chinese Academy of SciencesBeijingChina Department of Computer and Information Science University of MacaoMacaoChina

Redundancy elimination techniques are extensively investigated to reduce storage overheads for cloud-assisted health *** eliminates the redundancy of duplicate blocks by storing one physical instance referenced by multiple *** compression is usually regarded as a complementary technique to deduplication to further remove the redundancy of similar blocks,but our observations indicate that this is disobedient when data have sparse duplicate *** addition,there are many overlapped deltas in the resemblance detection process of post-deduplication delta compression,which hinders the efficiency of delta compression and the index phase of resemblance detection inquires abundant non-similar blocks,resulting in inefficient system ***,a multi-feature-based redundancy elimination scheme,called MFRE,is proposed to solve these *** similarity feature and temporal locality feature are excavated to assist redundancy elimination where the similarity feature well expresses the duplicate ***,similarity-based dynamic post-deduplication delta compression and temporal locality-based dynamic delta compression discover more similar base blocks to minimise overlapped deltas and improve compression ***,the clustering method based on block-relationship and the feature index strategy based on bloom filters reduce IO overheads and improve system *** demonstrate that the proposed method,compared to the state-of-the-art method,improves the compression ratio and system throughput by 9.68%and 50%,respectively.

关键词： big data cloud computing compression data compression medical applications performance evaluation

来源：评论

学校读者我要写书评

暂无评论

Maximizing Resource Efficiency in Cloud Data centers through Knowledge-Based Flower Pollination Algorithm (KB-FPA)

引用

computers, Materials & Continua 2024年第6期79卷 3757-3782页

作者： Nidhika Chauhan Navneet Kaur Kamaljit Singh Saini Sahil Verma Kavita Ruba Abu Khurma Pedro A.Castillo University Institute of Computing Department Chandigarh UniversityPunjab140413India Department of Computer Science and Engineering Chandigarh UniversityPunjab140413India Universidade Federal do Piauí TeresinaPiauí64049-550Brazil MEU Research Unit Faculty of Information TechnologyMiddle East UniversityAmman11831Jordan Applied Science Research Center Applied Science Private UniversityAmman11931Jordan Department of Computer Engineering Automatics and RoboticsUniversity of GranadaGranada18071Spain

Cloud computing is a dynamic and rapidly evolving field,where the demand for resources fluctuates *** paper delves into the imperative need for adaptability in the allocation of resources to applications and services within cloud computing *** motivation stems from the pressing issue of accommodating fluctuating levels of user demand *** adhering to the proposed resource allocation method,we aim to achieve a substantial reduction in energy *** reduction hinges on the precise and efficient allocation of resources to the tasks that require those most,aligning with the broader goal of sustainable and eco-friendly cloud computing *** enhance the resource allocation process,we introduce a novel knowledge-based optimization *** this study,we rigorously evaluate its efficacy by comparing it to existing algorithms,including the Flower Pollination Algorithm(FPA),Spark Lion Whale Optimization(SLWO),and Firefly *** findings reveal that our proposed algorithm,Knowledge Based Flower Pollination Algorithm(KB-FPA),consistently outperforms these conventional methods in both resource allocation efficiency and energy consumption *** paper underscores the profound significance of resource allocation in the realm of cloud *** addressing the critical issue of adaptability and energy efficiency,it lays the groundwork for a more sustainable future in cloud computing *** contribution to the field lies in the introduction of a new resource allocation strategy,offering the potential for significantly improved efficiency and sustainability within cloud computing infrastructures.

关键词： Cloud computing resource allocation energy consumption optimization algorithm flower pollination algorithm

来源：评论

学校读者我要写书评

暂无评论

SELF-JOINT SUPERVISED LEARNING 10

SELF-JOINT SUPERVISED LEARNING

引用

10th International Conference on Learning Representations, ICLR 2022

作者： Kardan, Navid Hill, Mitchell Shah, Mubarak Center for Research in Computer Vision Department of Computer Science Department of Statistics and Data Science University of Central Florida United States

Supervised learning is a fundamental framework used to train machine learning systems. A supervised learning problem is often formulated using an i.i.d. assumption that restricts model attention to a single relevant signal at a time when predicting. This contrasts with the human ability to actively use related samples as reference when making decisions. We hypothesize that the restriction to a single signal for each prediction in the standard i.i.d. framework contributes to well-known drawbacks of supervised learning: making overconfident predictions and vulnerability to overfitting, adversarial attacks, and out-of-distribution data. To address these limitations, we propose a new supervised learning paradigm called self-joint learning that generalizes the standard approach by modeling the joint conditional distribution of two observed samples, where each sample is an image and its label. Rather than assuming samples are independent, our models explicitly learn the sample-to-sample relation of conditional independence. Our framework can naturally incorporate auxiliary unlabeled data to further improve the performance. Experiments on benchmark image datasets show our method offers significant improvement over standard supervised learning in terms of accuracy, robustness against adversarial attacks, out-of-distribution detection, and overconfidence mitigation. © 2022 ICLR 2022 - 10th International Conference on Learning Representationss. All rights reserved.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Object Detection using Patch Based Image Classifier and Gradient Importance Map

引用

International Journal of Information Technology (Singapore) 2025年第4期17卷 2407-2416页

作者： Jain, Vanita Pillai, Manu S. Jain, Achin Dubey, Arun Kumar Department of Electronic Science University of Delhi Delhi India Center for Research in Computer Vision University of Central Florida Orlando United States Bharati Vidyapeeth’s College of Engineering New Delhi India

Image classification in computer vision has seen tremendous amount of success in recent years. Deep learning has played a pivotal role in achieving human level performance in many image recognition challenges and benchmarks. Even though, image classification has been so successful, no other closely related domains have taken advantage from the efforts put into development of image classification methods. One such closely related field is of Object Detection. Object detection or localisation is a computer vision problem whose solutions have not been victorious enough to human level performance. Many challenges arise when developing object detection models for newly generated domains, one of which is labelling of datasets. Preparation of dataset is one of the most cumbersome and expensive task to accomplish while developing an object detection model. Although, image classifiers are used as a feature extractor in object detection training regimes, their localisation abilities are barely studied. In this paper, we propose an object detection training regime, that does not rely on bounding box labelled datasets, hence unsupervised in nature, and is solely based on trained image classifiers. We build up on our hypothesis, that, "if an image classifier is able to predict what object is in the input image, then it must have information about where the object is, we just need a mechanism to extract that information from it". Precisely, we divide the input image into patches of same size and employ a parameter restricted convolutional classifier on each patches to predict whether it contains the object or not, we call this our patch-based image classifier (the object here is the prediction of the trained image classifier). The training of the patch-based classifier is not straightforward as there is no true labels for each patches on which we can reduce the binary cross-entropy. Therefore, we propose a loss function weighted by the importance map, which we generate using Grad

关键词： Deep learning Grad-CAM Image classification Object detection

来源：评论

学校读者我要写书评

暂无评论

DUAL STUDENT NETWORKS FOR DATA-FREE MODEL STEALING 11

DUAL STUDENT NETWORKS FOR DATA-FREE MODEL STEALING

引用

11th International Conference on Learning Representations, ICLR 2023

作者： Beetham, James Kardan, Navid Mian, Ajmal Shah, Mubarak Center for Research in Computer Vision University of Central Florida OrlandoFL32816 United States Department of Computer Science University of Western Australia CrawleyWA6009 Australia

Data-free model stealing aims to replicate a target model without direct access to either the training data or the target model. To accomplish this, existing methods use a generator to produce samples in order to train a student model to match the target model outputs. To this end, the two main challenges are estimating gradients of the target model without access to its parameters, and generating a diverse set of training samples that thoroughly explores the input space. We propose a Dual Student method where two students are symmetrically trained in order to provide the generator a criterion to generate samples that the two students disagree on. On one hand, disagreement on a sample implies at least one student has classified the sample incorrectly when compared to the target model. This incentive towards disagreement implicitly encourages the generator to explore more diverse regions of the input space. On the other hand, our method utilizes gradients of student models to indirectly estimate gradients of the target model. We show that this novel training objective for the generator network is equivalent to optimizing a lower bound on the generator's loss if we had access to the target model gradients. In other words, our method alters the standard data-free model stealing paradigm by substituting the target model with a separate student model, thereby creating a lower bound which can be directly optimized without additional target model queries or separate synthetic datasets. We show that our new optimization framework provides more accurate gradient estimation of the target model and better accuracies on benchmark classification datasets. Additionally, our approach balances improved query efficiency with training computation cost. Finally, we demonstrate that our method serves as a better proxy model for transfer-based adversarial attacks than existing data-free model stealing methods. © 2023 11th International Conference on Learning Representations, ICLR 2023.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Multi-stage Conditional GAN Architectures for Person-Image Generation 1

引用

1st International Conference and 2nd International Conference on Deep Learning Theory and Applications, DeLTA 2020 and DeLTA 2021

作者： Kurupathi, Sheela Raju Dumpala, Veeru Stricker, Didier Department of Computer Science Technical University of Kaiserslautern Kaiserslautern Germany Augmented Vision German Research Center for Artificial Intelligence Kaiserslautern Germany

ISBN: (数字)9783031373206

ISBN: (纸本)9783031373190

Generating realistic human images has been of great value in recent times due to their varied application in robotics, computer Graphics, Movie Making, and Games. Advancements in Artificial Intelligence (AI) and Machine learning (ML) lead to the rapid growth of integrating every aspect into AI and ML. There are many deep learning models like Variational Auto Encoders (VAE), Stacked Hourglass networks and Generative Adversarial Networks (GANs) for solving human-image generation. However, it is still difficult for these models to generalize well to the real-world person-image generation task qualitatively. In this paper, we develop a multi-stage model based on Conditional GANs, which could synthesize the new image of the target person given the image of the person and the target (novel) pose desired in real-world scenarios. The model uses 2D keypoints to represent human poses. We propose a Multi-stage Person Generation (MPG) model, in which we have modified the Generator architecture of Pose Guided Person Image Generation (PG2) resulting in two approaches. The first three-stage person generation approach has an additional generator integrated to base architecture and has trained the model end-to-end. The second two-stage person generation approach has a novel texture feature block in stage-II and has been trained incrementally to improve the generation of human-images qualitatively. The proposed two-stage MPG approach has generated promising results on Market-1501 Dataset. The claims are supported by benchmarking the proposed models with recent state-of-the-art models. We also show how the multi-stage conditional GAN architectures influence the process of human-image synthesis. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Edge-SAN: An Edge-Prompted Foundation Model for Accurate Nuclei Instance Segmentation in Histology Images

Edge-SAN: An Edge-Prompted Foundation Model for Accurate Nuc...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Wu, Xuening Shen, Yiqing Zhao, Qing Kang, Yanlan Hu, Ruiqi Zhang, Wenqiang Fudan University Shanghai Engineering Research Center of Ai & Robotics Academy for Engineering & Technology Shanghai China Johns Hopkins University Department of Computer Science BaltimoreMD United States Hong Kong Polytechnic University Department of Applied Mathematics Hong Kong Fudan University Engineering Research Center of Ai & Robotics Ministry of Education Academy for Engineering & Technology School of Computer Science Shanghai China

ISBN: (纸本)9798350386226

Accurate nuclei segmentation is fundamental in histology image analysis, playing an essential role in cancer grading and diagnosis. However, this task remains challenging due to variations in staining protocols, heterogeneity among nuclei types, and the densely clustered nature of nuclei. While SAM exhibits zero-shot generalization capabilities in natural image segmentation, its performance degrades when applied to nuclei segmentation in histology images. Existing adaptations of SAM for medical imaging primarily focus on organ or lesion segmentation, which differs substantially from nuclei segmentation due to the unique characteristics of nuclei - specifically, their sparse distribution combined with dense clustering. To address these challenges, we propose Edge-SAN (Segment Any Nuclei with Edge Prompting), an interactive segmentation foundation model specifically designed for nuclei segmentation. Edge-SAN introduces a novel edge prompting method that enhances the delineation of nuclei boundaries, particularly among densely clustered nuclei, by leveraging edge information to improve segmentation accuracy. We evaluate Edge-SAN on 12 diverse datasets in both few-shot and zero-shot scenarios, demonstrating its effectiveness as a foundation model for nuclei segmentation, achieving 66.81% AJI and 73.13% DSC - improvements of 16.33% and 15.48% over SAM-Med2D, respectively. The code is available at https://***/deep-geo/Edge-SAN. © 2024 IEEE.

关键词： Deep Learning Digital Pathology Foundation Model Nuclei Segmentation Segment Anything Model

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：