检索结果-内蒙古大学图书馆

Effects of tDCS Associated with Proprioceptive Exercises on Postural Control in Individuals with Total Blindness

Gait & Posture 2024年 113卷 166-167页

作者： Rodolfo B Parreira Roberta C Toledo Deborah C S Cardoso Gabrielly R C Silva Carvalho P Milena C Giovana L Silva Iransé O Silva Manuela Galli Veronica Cimolin Claudia Oliveira Evangelical University of Goiás- UniEVANGÉLICA Postgraduate Program Anápolis Brazil Evangelical University of Goiás- UniEVANGÉLICA Undergraduate Course in Physiotherapy Anápolis Brazil Politecnico di Milano Department of Electronic- Information and Bioengineering Milan Italy São Paulo Santa Casa School of Medical Sciences Health Sciences Program São Paulo Brazil

来源：评论

学校读者我要写书评

暂无评论

FedPerfix: Towards Partial Model Personalization of Vision Transformers in Federated Learning

FedPerfix: Towards Partial Model Personalization of Vision T...

引用

International Conference on Computer Vision (ICCV)

作者： Guangyu Sun Matias Mendieta Jun Luo Shandong Wu Chen Chen Center for Research in Computer Vision University of Central Florida USA Intelligent Systems Program University of Pittsburgh USA Department of Radiology Biomedical Informatics and Bioengineering University of Pittsburgh USA

Personalized Federated Learning (PFL) represents a promising solution for decentralized learning in heterogeneous data environments. Partial model personalization has been proposed to improve the efficiency of PFL by selectively updating local model parameters instead of aggregating all of them. However, previous work on partial model personalization has mainly focused on Convolutional Neural Networks (CNNs), leaving a gap in understanding how it can be applied to other popular models such as Vision Transformers (ViTs). In this work, we investigate where and how to partially personalize a ViT model. Specifically, we empirically evaluate the sensitivity to data distribution of each type of layer. Based on the insights that the self-attention layer and the classification head are the most sensitive parts of a ViT, we propose a novel approach called FedPerfix, which leverages plugins to transfer information from the aggregated model to the local client as a personalization. Finally, we evaluate the proposed approach on CIFAR-100, OrganAMNIST, and Office-Home datasets and demonstrate its effectiveness in improving the model’s performance compared to several advanced PFL methods. Code is available at https://***/imguangyu/FedPerfix

关键词：

来源：评论

学校读者我要写书评

暂无评论

In vivo Detection of NO using Photonic Microring Resonators

In vivo Detection of NO using Photonic Microring Resonators

引用

CLEO: Science and Innovations, S and I 2022

作者： Hassan, Sakib Zhao, Xuan Schreib, Christian C. Duret, Guillaume Veiseh, Omid Robinson, Jacob T. Electrical & Computer Engineering Rice University HoustonTX77005 United States Applied Physics Graduate Program Smalley-Curl Institute Rice University HoustonTX77005 United States Department of Bioengineering Rice University HoustonTX77005 United States

Here we show how photonic microring resonators enable real-time detection of Nitric Oxide (NO) in vivo, which is critical for physiological monitoring particularly during the wound healing process. © Optica Publi... 详细信息

ISBN: (纸本)9781557528209

关键词： Nitric oxide

来源：评论

学校读者我要写书评

暂无评论

Novel Circuit Design for a 3D Position Sensitive Scintillation TOF-PET Detector that Achieves ~100 ps CTR

Novel Circuit Design for a 3D Position Sensitive Scintillati...

引用

2022 IEEE Nuclear Science Symposium, Medical Imaging Conference, and Room Temperature Semiconductor Detector Conference, IEEE NSS MIC RTSD 2022

作者： Pourashraf, Shirin Zhao, Zhixiang Innes, Derek Gonzalez-Montoro, Andrea Cates, Joshua W. Levin, Craig S. Stanford University Department of Radiology StanfordCA94305 United States Lawrence Berkeley National Laboratory Applied Nuclear Physics Program BerkeleyCA94720 United States Stanford University Molecular Imaging Program Departments of Radiology Bioengineering Physics and Electrical Engineering StanfordCA94305 United States The School of Biomedical Engineering Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9781665488723

We have built a scalable mixed-signal readout circuit for a 3D position sensitive PET detector that achieves ~100 ps FWHM CTR. For this work we used individual 3×3×3 mm3 and 3×3×10 mm3 LGSO crystals, as well as 2×4 arrays of 3×3×10 mm3 crystals side-coupled to arrays of 3.16×3.16 mm2 SiPMs. The preliminary results verify that the CPLD is able to resolve 3D interaction positions along the array with © 2022 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MultiWienerNet: Deep learning for fast shift-varying deconvolution

MultiWienerNet: Deep learning for fast shift-varying deconvo...

引用

Computational Optical Sensing and Imaging, COSI 2021 - Part of OSA Imaging and Applied Optics Congress 2021

作者： Shuai, Richard W. Yanny, Kyrollos Monakhova, Kristina Waller, Laura Department of Electrical Engineering and Computer Sciences University of California Berkeley United States UCB/UCSF Joint Graduate Program in Bioengineering University of California Berkeley United States

We present a deep-learning method based on Wiener filters and U-Nets that performs image reconstruction in systems with spatially-varying aberrations. We train on simulated microscopy measurements and test on experime... 详细信息

ISBN: (纸本)9781557528209

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Cell-mediated and cell membrane-coated nanoparticles for drug delivery and cancer therapy

引用

Cancer Drug Resistance 2020年第4期3卷 879-911页

作者： Serkan Yaman Uday Chintapula Edgar Rodriguez Harish Ramachandramoorthy Kytai T.Nguyen Department of Bioengineering University of Texas at ArlingtonArlingtonTX 76010USA Joint Bioengineering Program University of Texas Southwestern Medical CenterDallasTX 75235USA

Nanotechnology-based drug delivery platforms have been developed over the last two decades because of their favorable features in terms of improved drug bioavailability and *** recent advancement in nanotechnology platforms,this approach still falls short to meet the complexity of biological systems and diseases,such as avoiding systemic side effects,manipulating biological interactions and overcoming drug resistance,which hinders the therapeutic outcomes of the NP-based drug delivery *** address these issues,various strategies have been developed including the use of engineered cells and/or cell membrane-coated *** membrane receptor profiles and characteristics are vital in performing therapeutic functions,targeting,and homing of either engineered cells or cell membrane-coated nanocarriers to the sites of *** this context,we comprehensively discuss various cell-and cell membrane-based drug delivery approaches towards cancer therapy,the therapeutic potential of these strategies,and the limitations associated with engineered cells as drug carriers and cell membrane-associated drug ***,we review various cell types and cell membrane receptors for their potential in targeting,immunomodulation and overcoming drug resistance in cancer.

关键词： Cell membrane-based drug delivery cell-mediated drug delivery membrane receptors drug carriers cancer drug resistance nanoparticles

来源：评论

学校读者我要写书评

暂无评论

Longitudinal Mammogram Exam-based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks

arXiv

引用

arXiv 2024年

作者： Zhou, Zhengbo Hao, Degan Arefan, Dooman Zuley, Margarita Sumkin, Jules Wu, Shandong Intelligent Systems Program University of Pittsburgh PittsburghPA United States Department of Radiology University of Pittsburgh PittsburghPA United States Department of Biomedical Informatics Department of Bioengineering University of Pittsburgh PittsburghPA United States

In breast cancer detection and diagnosis, the longitudinal analysis of mammogram images is crucial. Contemporary models excel in detecting temporal imaging feature changes, thus enhancing the learning process over sequential imaging exams. Yet, the resilience of these longitudinal models against adversarial attacks remains underexplored. In this study, we proposed a novel attack method that capitalizes on the feature-level relationship between two sequential mammogram exams of a longitudinal model, guided by both cross-entropy loss and distance metric learning, to achieve significant attack efficacy, as implemented using attack transferring in a black-box attacking manner. We performed experiments on a cohort of 590 breast cancer patients (each has two sequential mammogram exams) in a case-control setting. Results showed that our proposed method surpassed several state-of-the-art adversarial attacks in fooling the diagnosis models to give opposite outputs. Our method remained effective even if the model was trained with the common defending method of adversarial training. © 2024, CC BY.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

MIXTURE OF EXPERTS MADE PERSONALIZED: FEDERATED PROMPT LEARNING FOR VISION-LANGUAGE MODELS

arXiv

引用

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

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：