检索结果-内蒙古大学图书馆

Crowdsourcing Skin Demarcations of Chronic Graft-Versus-Host Disease in Patient Photographs: Training Versus Performance Study

学校读者我要写书评

暂无评论

JMIR Dermatology

JMIR Dermatology 2023年第1期6卷 e48589页

作者： McNeil, Andrew J. Parks, Kelsey Liu, Xiaoqi Jiang, Bohan Coco, Joseph McCool, Kira Fabbri, Daniel Duhaime, Erik P. Dawant, Benoit M. Tkaczyk, Eric R. Dermatology Service and Research Service Department of Veterans Affairs Tennessee Valley Healthcare System Nashville TN United States Department of Dermatology Vanderbilt University Medical Center Nashville TN United States Department of Electrical and Computer Engineering Vanderbilt University Nashville TN United States Department of Biomedical Informatics Vanderbilt University Medical Center Nashville TN United States Centaur Labs Boston MA United States

Background: Chronic graft-versus-host disease (cGVHD) is a significant cause of long-term morbidity and mortality in patients after allogeneic hematopoietic cell transplantation. Skin is the most commonly affected organ, and visual assessment of cGVHD can have low reliability. Crowdsourcing data from nonexpert participants has been used for numerous medical applications, including image labeling and segmentation tasks. Objective: This study aimed to assess the ability of crowds of nonexpert raters-individuals without any prior training for identifying or marking cGHVD-to demarcate photos of cGVHD-affected skin. We also studied the effect of training and feedback on crowd performance. Methods: Using a Canfield Vectra H1 3D camera, 360 photographs of the skin of 36 patients with cGVHD were taken. Ground truth demarcations were provided in 3D by a trained expert and reviewed by a board-certified dermatologist. In total, 3000 2D images (projections from various angles) were created for crowd demarcation through the DiagnosUs mobile app. Raters were split into high and low feedback groups. The performances of 4 different crowds of nonexperts were analyzed, including 17 raters per image for the low and high feedback groups, 32-35 raters per image for the low feedback group, and the top 5 performers for each image from the low feedback group. Results: Across 8 demarcation competitions, 130 raters were recruited to the high feedback group and 161 to the low feedback group. This resulted in a total of 54,887 individual demarcations from the high feedback group and 78,967 from the low feedback group. The nonexpert crowds achieved good overall performance for segmenting cGVHD-affected skin with minimal training, achieving a median surface area error of less than 12% of skin pixels for all crowds in both the high and low feedback groups. The low feedback crowds performed slightly poorer than the high feedback crowd, even when a larger crowd was used. Tracking the 5 most reliabl

关键词： artificial intelligence cGVHD crowdsourcing dermatology feasibility graft-versus-host disease imaging labeling medical image segmentation skin

Can large language models provide useful feedback on research papers? A large-scale empirical analysis

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liang, Weixin Zhang, Yuhui Cao, Hancheng Wang, Binglu Ding, Daisy Yi Yang, Xinyu Vodrahalli, Kailas He, Siyu Smith, Daniel Scott Yin, Yian McFarland, Daniel A. Zou, James Department of Computer Science Stanford University StanfordCA94305 United States Kellogg School of Management Northwestern University EvanstonIL60208 United States Department of Biomedical Data Science Stanford University StanfordCA94305 United States Department of Information Science Cornell University IthacaNY14850 United States Department of Electrical Engineering Stanford University StanfordCA94305 United States Graduate School of Education Stanford University StanfordCA94305 United States

Expert feedback lays the foundation of rigorous research. However, the rapid growth of scholarly production and intricate knowledge specialization challenge the conventional scientific feedback mechanisms. High-quality peer reviews are increasingly difficult to obtain. Researchers who are more junior or from under-resourced settings have especially hard times getting timely feedback. With the breakthrough of large language models (LLM) such as GPT-4, there is growing interest in using LLMs to generate scientific feedback on research manuscripts. However, the utility of LLM-generated feedback has not been systematically studied. To address this gap, we created an automated pipeline using GPT-4 to provide comments on the full PDFs of scientific papers. We evaluated the quality of GPT-4’s feedback through two large-scale studies. We first quantitatively compared GPT-4’s generated feedback with human peer reviewer feedback in 15 Nature family journals (3,096 papers in total) and the ICLR machine learning conference (1,709 papers). The overlap in the points raised by GPT-4 and by human reviewers (average overlap 30.85% for Nature journals, 39.23% for ICLR) is comparable to the overlap between two human reviewers (average overlap 28.58% for Nature journals, 35.25% for ICLR). The overlap between GPT-4 and human reviewers is larger for the weaker papers (i.e., rejected ICLR papers;average overlap 43.80%). We then conducted a prospective user study with 308 researchers from 110 US institutions in the field of AI and computational biology to understand how researchers perceive feedback generated by our GPT-4 system on their own papers. Overall, more than half (57.4%) of the users found GPT-4 generated feedback helpful/very helpful and 82.4% found it more beneficial than feedback from at least some human reviewers. While our findings show that LLM-generated feedback can help researchers, we also identify several limitations. For example, GPT-4 tends to focus on certain aspects

关键词： Quality control

De-speckling of optical coherence tomography images using a generative adversarial network

学校读者我要写书评

暂无评论

De-speckling of optical coherence tomography images using a ...

Optical Coherence Tomography, OCT 2020

作者： Dong, Zhao Ni, Guangming Liu, Guoyan Jerwick, Jason Zhou, Chao Department of Biomedical Engineering Washington University in St Louis 1 Brookings Dr. St. LouisMO63105 United States Department of Electrical and Computer Engineering Lehigh University 27 Memorial Drive W BethlehemPA18015 United States Department of Dermatology Affiliated Hospital of Weifang Medical University 465 Yuhe Rd Weifang261041 China

We developed a custom generative adversarial network (GAN) that can effectively reduce speckle noise in OCT images and videos without increase system complexity while maintaining spatial and temporal resolutions. ... 详细信息

ISBN: (纸本)9781943580743

关键词： Generative adversarial networks

Towards frequency adaptation for delayed feedback deep brain stimulations

学校读者我要写书评

暂无评论

Neural Regeneration Research 2018年第3期13卷 408-409页

作者： Mohammad Daneshzand Miad Faezipour Buket D.Barkana D-BEST Lab Departments of Computer Science and Engineering and Biomedical Engineering University of Bridgeport Department of Electrical Engineering University of Bridgeport

In neurodegenerative disorders such as Parkinson＇s disease （PD）, deep brain stimulation （DBS） is a desirable approach when the medication is less effective for treating the symptoms. DBS incorporates transferring electrical pulses to a specific tissue of the central nervous system, obtaining therapeutic results by modulating the neuronal activity of that region. DBS has certain advantages such as reversibility and adjustability features over medication, since the neuronal firing patterns can be recorded and used to alter the parameters of the DBS signal （Benabid et al., 2009）. One of the DBS indications is its ability to suppress the abnormal neuronal activity to treat symptoms like tremor, akinesia and dystonia.

关键词： Figure DBS Towards frequency adaptation for delayed feedback deep brain stimulations STN HFS

FlowReg: Fast deformable unsupervised medical image registration using optical flow

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Mocanu, Sergiu Moody, Alan R. Khademi, April Electrical Computer and Biomedical Engineering Ryerson University TorontoON Canada Department of Medical Imaging University of Toronto TorontoON Canada Keenan Research Center for Biomedical Science St. Michael’s Hospital Unity Health Network TorontoON Canada a partnership between St. Michael’s Hospital Ryerson University TorontoON Canada

In this work we propose FlowReg, a deep learning-based framework that performs unsupervised image registration for neuroimaging applications. The system is composed of two architectures that are trained sequentially: FlowReg-A which affinely corrects for gross differences between moving and fixed volumes in 3D followed by FlowReg-O which performs pixel-wise deformations on a slice-by-slice basis for fine tuning in 2D. FlowReg-A warps the moving volume using gross global parameters to align rotation, scale, shear, and translation to the fixed volume. A correlation loss that encourages global alignment between the moving and the fixed volumes is employed to regress the affine parameters. The deformable network FlowReg-O operates on 2D image slices and is based on the optical flow CNN network that is adapted to neuroimaging with three loss components. The photometric loss minimizes pixel intensity differences, the smoothness loss encourages similar magnitudes between neighbouring vectors, and a correlation loss that is used to maintain the intensity similarity between fixed and moving image slices. The proposed method is compared to four open source registration techniques ANTs, Demons, SE, and Voxelmorph for FLAIR MRI applications. In total, 4643 FLAIR MR imaging volumes (approximately 255, 000 image slices) are used from dementia and vascular disease cohorts, acquired from over 60 international centres with varying acquisition parameters. To quantitatively assess the performance of the registration tools, a battery of novel validation metrics are proposed that focus on the structural integrity of tissues, spatial alignment, and intensity similarity. Experimental results show FlowReg (FlowReg-A+O) performs better than iterative-based registration algorithms for intensity and spatial alignment metrics with a Pixelwise Agreement (PWA) of 0.65, correlation coefficient (R) of 0.80, and Mutual Information (MI) of 0.29. Among the deep learning frameworks evaluated, FlowReg-

关键词： Optical flows

Applications of deep learning techniques for automated multiple Sclerosis detection using magnetic resonance imaging: A review

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shoeibi, Afshin Khodatars, Marjane Jafari, Mahboobeh Moridian, Parisa Rezaei, Mitra Alizadehsani, Roohallah Khozeimeh, Fahime Gorriz, Juan Manuel Heras, Jónathan Panahiazar, Maryam Nahavandi, Saeid Acharya, U. Rajendra K. N. Toosi University of Technology Tehran Iran Faculty of Engineering Mashhad Branch Islamic Azad University Mashhad Iran Electrical and Computer Engineering Faculty Semnan University Semnan Iran Faculty of Engineering Science and Research Branch Islamic Azad University Tehran Iran Electrical and Computer Engineering Dept. Tarbiat Modares University Tehran Iran Deakin University Geelong Australia Department of Signal Theory Networking and Communications Universidad de Granada Spain Department of Psychiatry University of Cambridge United Kingdom Department of Mathematics and Computer Science University of La Rioja La Rioja Spain University of California San Francisco San FranciscoCA United States Department of Biomedical Engineering School of Science and Technology Singapore University of Social Sciences Singapore Singapore Dept. of Electronics and Computer Engineering Ngee Ann Polytechnic Singapore599489 Singapore Department of Bioinformatics and Medical Engineering Asia University Taiwan

Multiple Sclerosis (MS) is a type of brain disease which causes visual, sensory, and motor problems for people with a detrimental effect on the functioning of the nervous system. In order to diagnose MS, multiple screening methods have been proposed so far;among them, magnetic resonance imaging (MRI) has received considerable attention among physicians. MRI modalities provide physicians with fundamental information about the structure and function of the brain, which is crucial for the rapid diagnosis of MS lesions. Diagnosing MS using MRI is time-consuming, tedious, and prone to manual errors. Research on the implementation of computer aided diagnosis system (CADS) based on artificial intelligence (AI) to diagnose MS involves conventional machine learning and deep learning (DL) methods. In conventional machine learning, feature extraction, feature selection, and classification steps are carried out by using trial and error;on the contrary, these steps in DL are based on deep layers whose values are automatically learn. In this paper, a complete review of automated MS diagnosis methods performed using DL techniques with MRI neuroimaging modalities is provided. Initially, the steps involved in various CADS proposed using MRI modalities and DL techniques for MS diagnosis are investigated. The important preprocessing techniques employed in various works are analyzed. Most of the published papers on MS diagnosis using MRI modalities and DL are presented. The most significant challenges facing and future direction of automated diagnosis of MS using MRI modalities and DL techniques are also provided. Copyright © 2021, The Authors. All rights reserved.

关键词： Magnetic resonance imaging

The Role of Frontal and Occipital Cortices in Processing Sustained Visual Attention in Young Adults with Attention-Deficit/Hyperactivity Disorder:A Functional Near-Infrared Spectroscopy Study

学校读者我要写书评

暂无评论

Neuroscience Bulletin 2020年第6期36卷 659-663页

作者： Ziyan Wu Yuyang Luo Yu Gao Ying Han Kai Wu Xiaobo Li Department of Electrical and Computer Engineering New Jersey Institute of TechnologyNewarkNJ 07102USA Department of Biomedical Engineering New Jersey Institute of TechnologyNewarkNJ 07102USA Department of Psychology Brooklyn College and the Graduate Center of the City University of New YorkBrooklynNY 10004USA Department of Neurology XuanWu Hospital of Capital Medical UniversityBeijing 100053China Beijing Institute of Geriatrics XuanWu Hospital of Capital Medical UniversityBeijing 100053China National Clinical Research Center for Geriatric Disorders Beijing 100053China Center of Alzheimer’s Disease Beijing Institute for Brain DisordersBeijing 100069China Department of Biomedical Engineering School of Materials Science and EngineeringSouth China University of TechnologyGuangzhou 510003China Guangzhou Huiai Hospital The Affiliated Brain Hospital of Guangzhou Medical UniversityGuangzhou 510370China Guangdong Engineering Technology Research Center for Diagnosis and Rehabilitation of Dementia Guangzhou 510500China

Dear Editor,Attention deficit/hyperactivity disorder(ADHD),characterized by age-inappropriate symptoms of inattention,hyperactivity,and impulsivity,is a highly prevalent and heritable childhood-onset neurodevelopmental disorder,with impairing symptoms that persist into adulthood in up to 65%of patients[1].Adults with persistent ADHD symptoms since childhood have been found to have neurocognitive impairments in multiple domains,especially in attention and cognitive control,which significantly contribute to an increased risk of social disability,educational and occupational failure,and other comorbid psychiatric disorders,resulting in significant economic burdens to the affected individuals, their families, andsociety [2, 3].

关键词： Corti impairment Infrared

Spatio-temporal dynamics of malaria in Rwanda between 2012 and 2022: a demography-specific analysis

学校读者我要写书评

暂无评论

Infectious Diseases of Poverty 2024年第5期13卷 100-100页

作者： Felix K.Rubuga Paula Moraga Ayman Ahmed Emmanuel Siddig Eric Remera Giovenale Moirano Guéladio Cissé Jürg Utzinger Swiss Tropical and Public Health Institute Allschwil Switzerland University of Basel Basel Switzerland College of Medicine and Health Sciences University of Rwanda Kigali Rwanda Center for Impact Innovation and Capacity building for Health Information systems and Nutrition (CIIC-HIN) Kigali Rwanda Institute of Endemic Diseases University of Khartoum Khartoum Sudan Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal Saudi Arabia Department of Medical Microbiology and Infectious Diseases Erasmus Medical Center University Medical Center Rotterdam Rotterdam The Netherlands Faculty of Medical Laboratory Sciences University of Khartoum Khartoum Sudan Rwanda Biomedical Center Kigali Rwanda Department of Medical Science University of Turin Torino Italy Barcelona Supercomputing Center Barcelona Spain

Background Despite global efforts to reduce and eventually interrupt malaria transmission,the disease remains a pressing public health problem,especially in sub-Saharan *** study presents a detailed spatio-temporal analysis of malaria transmission in Rwanda from 2012 to *** main objective was to gain insights into the evolving patterns of malaria and to inform and tailor effective public health *** The study used yearly aggregated data of malaria cases from the Rwanda health management information *** employed a multifaceted analytical approach,including descriptive statistics and spatio-temporal analysis across three demographic groups:children under the age of 5 years,and males and females above 5 *** spatially explicit models and spatio scan statistics were utilised to examine geographic and temporal patterns of relative risks and to identify clusters of malaria *** We observed a significant increase in malaria cases from 2014 to 2018,peaking in 2016 for males and females aged above 5 years with counts of 98,645 and 116,627,respectively and in 2018 for under 5-year-old children with 84,440 cases with notable geographic *** like Kamonyi(Southern Province),Ngoma,Kayonza and Bugesera(Eastern Province)exhibited high burdens,possibly influenced by factors such as climate,vector control practices,and cross-border *** spatially explicit modeling revealed elevated relative risks in numerous districts,underscoring the heterogeneity of malaria transmission in these districts,and thus contributing to an overall rising trend in malaria cases until 2018,followed by a subsequent *** findings emphasize that the heterogeneity of malaria transmission is potentially driven by ecologic,socioeconomic,and behavioural *** The study underscores the complexity of malaria transmission in Rwanda and calls for climate adaptive,gender-,age-and district-specific strategies in

关键词： Malaria transmission Epidemiology Public health Rwanda Spatio-temporal analysis Rwanda

XCloud-pFISTA: A Medical Intelligence Cloud for Accelerated MRI

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhou, Yirong Qian, Chen Guo, Yi Wang, Zi Wang, Jian Qu, Biao Guo, Di You, Yongfu Qu, Xiaobo Biomedical Intelligent Cloud R&D Center Department of Electronic Science National Institute for Data Science in Health and Medicine Xiamen University Xiamen361005 China Department of Instrumental and Electrical Engineering Xiamen University Xiamen361005 China School of Computer and Information Engineering Xiamen University of Technology Xiamen361024 China China Mobile Group Xiamen361005 China Biomedical Intelligent Cloud R&D Center School of Electronic Science and Engineering Xiamen University Xiamen361005 China

— Machine learning and artificial intelligence have shown remarkable performance in accelerated magnetic resonance imaging (MRI). Cloud computing technologies have great advantages in building an easily accessible platform to deploy advanced algorithms. In this work, we develop an open-access, easy-to-use and high-performance medical intelligence cloud computing platform (XCloud-pFISTA) to reconstruct MRI images from undersampled k-space data. Two state-of-the-art approaches of the Projected Fast Iterative Soft-Thresholding Algorithm (pFISTA) family have been successfully implemented on the cloud. This work can be considered as a good example of cloud-based medical image reconstruction and may benefit the future development of integrated reconstruction and online diagnosis system. Copyright © 2021, The Authors. All rights reserved.

关键词： Magnetic resonance imaging