检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Luo, Jun Chen, Chen Wu, Shandong Intelligent Systems Program University of Pittsburgh PittsburghPA15213 United States Intelligent Systems Program Department of Radiology Department of Biomedical Informatics Department of Bioengineering University of Pittsburgh PittsburghPA15213 United States Center for Research in Computer Vision University of Central Florida OrlandoFL32816 United States

Federated prompt learning benefits federated learning with CLIP-like Vision-Language Model’s (VLM’s) robust representation learning ability through prompt learning. However, current federated prompt learning methods are habitually restricted to the traditional FL paradigm, where the participating clients are generally only allowed to download a single globally aggregated model from the server. While justifiable for training full-sized models under federated settings, in this work, we argue that this paradigm is ill-suited for lightweight prompts. By facilitating the clients to download multiple pre-aggregated prompts as fixed non-local experts, we propose Personalized Federated Mixture of Adaptive Prompts (pFedMoAP), a novel FL framework that personalizes the prompt learning process through the lens of Mixture of Experts (MoE). pFedMoAP implements a local attention-based gating network that learns to generate enhanced text features for better alignment with local image data, benefiting from both local and downloaded non-local adaptive prompt experts. Extensive experiments on 9 datasets under various federated settings demonstrate the efficacy of the proposed pFedMoAP algorithm. The code is available at https://***/ljaiverson/pFedMoAP. © 2024, CC BY.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Empirical influence functions to understand the logic of fine-tuning

arXiv

引用

arXiv 2024年

作者： Matelsky, Jordan K. Ungar, Lyle Kording, Konrad P. Department of Bioengineering University of Pennsylvania PhiladelphiaPA United States Research & Exploratory Development Department Johns Hopkins University Applied Physics Laboratory LaurelMD United States Department of Computer and Information Science University of Pennsylvania PhiladelphiaPA United States Department of Bioengineering Department of Neuroscience University of Pennsylvania PhiladelphiaPA United States CIFAR LMB Program Canada

Understanding the process of learning in neural networks is crucial for improving their performance and interpreting their behavior. This can be approximately understood by asking how a model's output is influenced when we fine-tune on a new training sample. There are desiderata for such influences, such as decreasing influence with semantic distance, sparseness, noise invariance, transitive causality, and logical consistency. Here we use the empirical influence measured using fine-tuning to demonstrate how individual training samples affect outputs. We show that these desiderata are violated for both for simple convolutional networks and for a modern LLM. We also illustrate how prompting can partially rescue this failure. Our paper presents an efficient and practical way of quantifying how well neural networks learn from fine-tuning stimuli. Our results suggest that popular models cannot generalize or perform logic in the way they appear to. © 2024, CC BY.

关键词： Sampling

来源：评论

学校读者我要写书评

暂无评论

Bioceramic-mediated chondrocyte hypertrophy promotes calcified cartilage formation for rabbit osteochondral defect repair

引用

Bioactive Materials 2024年第10期40卷 306-317页

作者： Rachel H.Koh Junhee Kim Jeong-Uk Kim Seunghyun L.Kim Arun Kumar Rajendran Seunghun SLee Heesoo Lee Joo Hyun Kim Ji Hoon Jeong Yongsung Hwang Jong Woo Bae Nathaniel S.Hwang School of Chemical and Biological Engineering Institute of Chemical ProcessesSeoul National UniversitySeoul08826South Korea Interdisciplinary Program in Bioengineering Seoul National UniversitySeoul08826South Korea Department of Biomedical Engineering Dongguk UniversitySeoul10326South Korea Soonchunhyang Institute of Medi-bio Science(SIMS) Soonchunhyang UniversityCheonan31151South Korea Department of Integrated Biomedical Science Soonchunhyang UniversityAsan31538South Korea Department of Orthopaedic Surgery Konkuk University Chungju HospitalKonkuk University School of MedicineChungju27376South Korea BioMAX Institute Seoul National UniversitySeoul08826South Korea

Osteochondral tissue is a highly specialized and complex tissue composed of articular cartilage and subchondral bone that are separated by a calcified cartilage *** or gradient scaffolds,often in conjunction with stem cells and growth factors,have been developed to mimic the respective layers for osteochondral defect *** this study,we designed a hyaline cartilage-hypertrophic cartilage bilayer graft(RGD/RGDW)with ***,we demonstrated that RGD peptide-modified chondroitin sulfate cryogel(RGD group)is chondro-conductive and capable of hyaline cartilage ***,we incorporated whitlockite(WH),a Mg^(2+)-containing calcium phosphate,into RGD cryogel(RGDW group)to induce chondrocyte hypertrophy and form collagen X-rich hypertrophic *** is the first study to use WH to produce hypertrophic ***-laden RGDW cryogel exhibited significantly upregulated expression of hypertrophy markers in vitro and formed ectopic hypertrophic cartilage in vivo,which mineralized into calcified cartilage in bone ***,RGD cryogel and RGDW cryogel were combined into bilayer(RGD/RGDW group)and implanted into rabbit osteochondral defect,where RGD layer supports hyaline cartilage regeneration and bioceramic-containing RGDW layer promotes calcified cartilage *** the RGD group(monolayer)formed hyaline-like neotissue that extends into the subchondral bone,the RGD/RGDW group(bilayer)regenerated hyaline cartilage tissue confined to its respective layer and promoted osseointegration for integrative defect repair.

关键词： Bioceramic Whitlockite Bilayer scaffold Chondrocyte hypertrophy Osteochondral defect repair

来源：评论

学校读者我要写书评

暂无评论

Security approaches for electronic health data handling through the Semantic Web: A scoping review

引用

SEMANTIC WEB 2023年第4期14卷 771-784页

作者： Lima, Vinicius Costa Alves, Domingos Bernardi, Filipe Andrade Rijo, Rui Pedro Charters Lopes Univ Sao Paulo Ribeirao Preto Med Sch Hlth Intelligence Lab Sao Paulo Brazil Univ Sao Paulo Sch Engn Sao Carlos Bioengineering Post Grad Program Sao Paulo Brazil Univ Sao Paulo Ribeirao Preto Med Sch Dept Social Med Sao Paulo Brazil Polytechn Inst Leiria Sch Technol & Management Leiria Portugal Inst Syst Engn & Computers Coimbra Portugal Ctr Hlth Technol & Serv Res Porto Portugal

Integration of health information systems are crucial to advance the effective delivery of healthcare for individuals and communities across organizational boundaries. Semantic Web technologies may be used to connect, correlate, and integrate heterogeneous datasets spread over the internet. However, when working with sensitive data, such as health data, security mechanisms are needed. A scoping review of the literature was undertaken to provide a broad view of security mechanisms applied to, or along with, Semantic Web technologies that could allow its use with health data. Searches were conducted in the most relevant databases for the scope of this work. The findings were classified according to the main objective and features presented by each solution. Twenty-six studies were included in the review. They introduced mechanisms that addressed several security attributes, such as authentication, authorization, integrity, availability, confidentiality, privacy, and provenance. These mechanisms support access control frameworks, semantic and functional interoperability infrastructures, and privacy compliance solutions. The findings suggest that the application and use of Semantic Web technologies is still growing, with the healthcare area being particularly interested. The main security mechanisms for SemanticWeb technologies, the key security attributes and properties, and the main gaps in the literature were identified, helping to understand the technical needs to mitigate the risks of handling personal health information over the SemanticWeb. Also, this research has shown that complex and robust solutions are available to successfully address several security properties and features, depending on the context that the electronic health data is being managed.

关键词： Semantic Web health information systems electronic health records computer security interoperability

来源：评论

学校读者我要写书评

暂无评论

Durable CRISPR-Based Epigenetic Silencing

引用

BioDesign Research 2021年第1期2021卷 39-46页

作者： Muneaki Nakamura Alexis E.Ivec Yuchen Gao Lei S.Qi Department of Bioengineering StanfordCA 94305USA Program in Human Biology StanfordCA 94305USA Cancer Biology Program StanfordCA 94305USA Department of Chemical and Systems Biology StanfordCA 94305USA ChEM-H Institute StanfordCA 94305USA

Development of CRISPR-based epigenome editing tools is important for the study and engineering of biological ***,we describe the design of a reporter system for quantifying the ability of CRISPR epigenome editors to produce a stable gene *** characterize the dynamics of durable gene silencing and reactivation,as well as the induced epigenetic changes of this *** report the creation of single-protein CRISPR constructs bearing combinations of three epigenetic editing domains,termed KAL,that can stably repress the gene *** system should allow for the development of novel epigenome editing tools which will be useful in a wide array of biological research and engineering applications.

关键词： expression behavior CRISPR

来源：评论

学校读者我要写书评

暂无评论

Temperature-sensitive polymeric nanogels encapsulating withβ-cyclodextrin and ICG complex for high-resolution deep-tissue ultrasound-switchable fluorescence imaging

引用

Nano Research 2020年第4期13卷 1100-1110页

作者： Ruilin Liu Tingfeng Yao Yang Liu Shuai Yu Liqin Ren Yi Hong Kytai T.Nguyen Baohong Yuan Ultrasound and Optical Imaging Laboratory Department of Bioengineeringthe University of Texas at ArlingtonArlingtonTX 76019USA Joint Biomedical Engineering Program the University of Texas at Arlington and the University of Texas Southwestern Medical CenterDallasTX 75390USA Department of Bioengineering the University of Texas at ArlingtonArlingtonTX 76019USA

One of the thorny problems currently impeding the applications of the fluorescence imaging technique is the poor spatial resolution in deep ***-switchable fluorescence(USF)imaging is a novel imaging tool that has recently been explored to possibly surmount the above-mentioned ***,αβ-cyclodextrin/indocyanine green(ICG)complex-encapsulated poly(N-isopropylacrylamide)(PNIPAM)nanogel was synthesized and studied for ex vivo/in vivo deep tissue/high-resolution near infrared USF(NIR-USF)*** be specific,our results revealed that the average diameter of the as-prepared nanogels was significantly decreased to-32 nm from-335 nm compared to the reported ICG-PNIPAM ***,the excitation/emission characteristics of the ICG itself in present nanogels were almost completely retained,and the resultant nanogel exhibited high physiological stability and positive *** particular,the signal-to-noise ratio of the USF image for the PNIPAM/P-cyclodextrin/ICG nanogel(33.01±2.42 dB)was prominently higher than that of the ICG-PNIPAM nanoparticles(18.73±0.33 dB)in 1.5-cm-thick chicken breast *** NIR-USF imaging in 3.5-cm-thick chicken breast tissues was achieved using this new *** e x v iv o NIR-USF imaging of the mouse liver was also successfully *** experiments showed that the present nanogels were able to be effectively accumulated into U87 tumor-bearing mice via enhanced permeability and retention effects,and the high-resolution NIR-USF imaging of in v ivo tumor was efficiently *** metabolism and in vivo biodistribution of the nanogels were ***,the results suggest that the current nanogel is a highly promising NIR-USF probe for deep tissue and high-resolution USF imaging.

关键词： deep tissue imaging high-resolution ultrasound-switchable NIR fluorescence imaging in vivo imaging

来源：评论

学校读者我要写书评

暂无评论

Quality analysis and study of tuberculosis diagnostic data

引用

Procedia Computer Science 2024年 239卷 939-946页

作者： Ana Clara de Andrade Mioto Pedro Emilio Andrade Martins Gabriel Modina Domingos Alves Vinicius Costa Lima Mariane Neiva Filipe Andrade Bernardi Bioengineering Postgraduate Program University of São Paulo São Carlos Brazil Ribeirão Preto Medical School University of São Paulo Ribeirão Preto Brazil Institute of Mathematical and Computer Sciences University of São Paulo Ribeirão Preto Brazil

According to the World Health Organization (WHO), annually, 10 million people suffer from Tuberculosis (TB) worldwide, an infection caused by the bacteria Mycobacterium tuberculosis. Clinical data is complex and characterized by heterogeneity, making the interoperability of systems challenging and impairing the availability, integration, and quality of data, essential for health activity. This negatively impacts both patient treatment and decision-making. At the data level, the identification and evaluation of the best quality attributes of clinical data and recommendations available in different articles explored during the article, will be carried out. The analysis will also look for data clusters that do not meet these recommendations, in order to identify unknown patterns that may help to understand the mechanisms behind the data quality and, consequently, the expected results for the evaluated project.

关键词： Tuberculosis Quality Analysis FAIR

来源：评论

学校读者我要写书评

暂无评论

OVERVIEW OF PROCESSING TECHNIQUES FOR SURFACE ELECTROMYOGRAPHY SIGNALS

arXiv

引用

arXiv 2023年

作者： Manjarres-Triana, Alejandra Acevedo-Serna, Juan Ramírez-Duque, Andrés A. Jiménez, Mario F. Pulido-Herrera, Edith Villarejo Mayor, John J. Bioengineering Program Universidad El Bosque Bogotá Colombia School of Computing Science University of Glasgow Glasgow United Kingdom School of Engineering Science and Technology Universidad del Rosario Bogotá Colombia Postgraduation Program of Physical Education Federal University of Parana Curitiba Brazil

Surface electromyography (sEMG) is a technology to assess muscle activation, which is an important component in applications related to diagnosis, treatment, progression assessment, and rehabilitation of specific individuals’ conditions. Recently, sEMG potential has been shown, since it can be used in a non-invasive manner;nevertheless, it requires careful signal analysis to support health professionals reliably. This paper briefly described the basic concepts involved in the sEMG, such as the physiology of the muscles, the data acquisition, the signal processing techniques, and classification methods that may be used to identify disorders or signs of abnormalities according to muscular patterns. Specifically, classification methods encompass digital signal processing techniques and machine learning with high potential in the field. We hope that this work serves as an introduction to researchers interested in this *** Codes 92C55 (Primary), 94A99 (secondary) © 2023, CC BY.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

EDM: Enhancing Quality Diversity of Medical Image Augmentation Under the User-specified Complex and Various Conditions

EDM: Enhancing Quality Diversity of Medical Image Augmentati...

引用

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： S. Y. Kim H. B. Chung J. S. Lee M. J. Kang Interdisciplinary Program in Artificial Intelligence Seoul National University Seoul Republic of Korea Seoul National University Nuclear Medicine Bioengineering and Biomedical Sciences Seoul Republic of Korea Seoul National University Mathematical Sciences Seoul Republic of Korea

ISBN: (数字)9798350388152

ISBN: (纸本)9798350388169

In the medical domain, the scarcity of data leads to class imbalances and fairness issues, significantly impairing deep learning performance and leading to the underrepresentation of minority patient groups. To address these challenges, data augmentation techniques employing Generative Adversarial Networks (GANs) and diffusion models have been extensively explored, with diffusion models particularly noted for their ability to enhance data diversity and quality. Despite their achievements, it remains a challenging task to generate a diverse range of samples from limited datasets and to achieve precise generation of images that align with user *** overcome these limitations, we propose a framework, EDM (Enhancing Quality Diversity of Medical image augmentation under the user-specified complex and various conditions), designed to generate diverse high-quality medical images under the various and complex multimodal conditions, representing age, sex, and diseases. Our framework comprises four steps. First, we propose a metric to quantize the necessity of data augmentation. Second, we prepare data augmentation. We generate medical sketch-image-text triplets using DiffSketcher from the real patients’ dataset. Then we employ these triplets to train ControlNet, enabling the generation of synthetic medical images from sketch data. Third, we generate synthetic medical images. We obtain multiple sketches for each image from a diversity-enhanced DiffSketcher, incorporating similarity loss. Using these sketches, we augment data from the fine-tuned ControlNet from the second step. Lastly, we introduce a metric to measure the diversity of the generated images, ensuring the effectiveness of our augmentation *** preliminary experimental results imply EDM can be the promising solution. Future experiments are planned to construct synthetic datasets of various modalities such as PET and CT for underrepresented patient groups.

关键词： Semiconductor device measurement Microwave integrated circuits Semiconductor detectors Data augmentation Diffusion models Generative adversarial networks Image augmentation

来源：评论

学校读者我要写书评

暂无评论

Accelerated Direct PET Attenuation and Scatter Corrections Using Generation-Prior Diffusion Model (GPDM)

Accelerated Direct PET Attenuation and Scatter Corrections U...

引用

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： M.J. Cho H.S. Shim J.S. Lee Interdisciplinary Program of Bioengineering Seoul National University Seoul Republic of Korea Integrated Major in Innovative Medical Science Seoul National University Seoul Republic of Korea Department of Nuclear Medicine Seoul National University Seoul Republic of Korea

ISBN: (数字)9798350388152

ISBN: (纸本)9798350388169

Attenuation and scatter corrections play pivotal roles in PET image processing for accurate quantification. While CT-based corrections are conventional, they encounter limitations such as respiratory motion and metal artifacts. Recent advances in generative deep neural networks offer promising alternatives by directly estimating attenuation and scatter-corrected (ASC) PET from non-attenuation and non-scatter-corrected (NASC) PET. In GAN-based methods, the generator typically learns to map input data to the desired output (e.g. NASC PET to ASC PET) in one step. While this approach can be effective in generating realistic images, it often encounters significant challenges, such as unstable training and mode collapse. On the other hand, Denosing Diffusion Probabilistic Model (DDPM), a recent innovative solution based on probability theory, has emerged to address these issues, and gradually refines the image over numerous steps (e.g., 1,000 steps). In our previous study, a conditional DDPM (cDDPM) model, where the diffusion process was conditioned with NASC PET to generate ASC PET, outperformed conventional CycleGAN and conditional GAN (cGAN) approaches in the direct PET attenuation and scatter corrections [1]. However, the high computational complexity is one of the main drawbacks of DDPM. The training and inference steps of DDPM can be computationally intensive, requiring significant resources in terms of time, memory, and hardware. To overcome these problems, this study proposes a new acceleration method for the fast learning and sampling process of DDPM. In this method, named Generation-Prior Diffusion Model (GPDM), cDDPM is conditioned with prior variables obtained by applying a U-NET or cGAN to generate ASC PET from NASC PET. GPDM can significantly reduce the sampling steps owing to the high similarity between conditioned images and ASC PET. Consequently, computational complexity can be effectively decreased.

关键词： Training Microwave integrated circuits Semiconductor detectors Image processing Memory management Metals Attenuation

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：