检索结果-内蒙古大学图书馆

The Family Level Assessment of Screen Use-Mobile Approach: Development of an Approach to Measure Children's Mobile Device Use

引用

JMIR Formative Research 2022年第10期6卷 e40452页

作者： Perez, Oriana Vadathya, Anil Kumar Beltran, Alicia Barnett, R. Matthew Hindera, Olivia Garza, Tatyana Musaad, Salma M. Baranowski, Tom Hughes, Sheryl O. Mendoza, Jason A. Sabharwal, Ashutosh Veeraraghavan, Ashok United States Department of Agriculture Agricultural Research Service Children's Nutrition Research Center Baylor College of Medicine Houston TX United States Department of Electrical & Computer Engineering Rice University Houston TX United States Center for Research Computing Rice University Houston TX United States Baylor College of Medicine Houston TX United States Public Health Sciences Division Fred Hutchinson Cancer Center Seattle WA United States Center for Child Health Behavior and Development Seattle Children's Research Institute Seattle WA United States

Background: There is a strong association between increased mobile device use and worse dietary habits, worse sleep outcomes, and poor academic performance in children. Self-report or parent-proxy report of children's screen time has been the most common method of measuring screen time, which may be imprecise or biased. Objective: The objective of this study was to assess the feasibility of measuring the screen time of children on mobile devices using the Family Level Assessment of Screen Use (FLASH)-mobile approach, an innovative method that leverages the existing features of the Android platform. Methods: This pilot study consisted of 2 laboratory-based observational feasibility studies and 2 home-based feasibility studies in the United States. A total of 48 parent-child dyads consisting of a parent and child aged 6 to 11 years participated in the pilot study. The children had to have their own or shared Android device. The laboratory-based studies included a standardized series of tasks while using the mobile device or watching television, which were video recorded. Video recordings were coded by staff for a gold standard comparison. The home-based studies instructed the parent-child dyads to use their mobile device as they typically use it over 3 days. Parents received a copy of the use logs at the end of the study and completed an exit interview in which they were asked to review their logs and share their perceptions and suggestions for the improvement of the FLASH-mobile approach. Results: The final version of the FLASH-mobile approach resulted in user identification compliance rates of >90% for smartphones and >80% for tablets. For laboratory-based studies, a mean agreement of 73.6% (SD 16.15%) was achieved compared with the gold standard (human coding of video recordings) in capturing the target child's mobile use. Qualitative feedback from parents and children revealed that parents found the FLASH-mobile approach useful for tracking how much time their chi

关键词： children mobile media apps mobile phone mobile phone use screen time tablet use

来源：评论

学校读者我要写书评

暂无评论

Deep learning applied to chest X-Rays: Exploiting and preventing shortcuts

arXiv

引用

arXiv 2020年

作者： Jabbour, Sarah Fouhey, David Kazerooni, Ella Sjoding, Michael W. Wiens, Jenna Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI United States Department of Radiology Department of Internal Medicine University of Michigan Ann ArborMI United States Institute for Healthcare Policy & Innovation Department of Internal Medicine University of Michigan Ann ArborMI United States

While deep learning has shown promise in improving the automated diagnosis of disease based on chest X-rays, deep networks may exhibit undesirable behavior related to shortcuts. This paper studies the case of spurious class skew in which patients with a particular attribute are spuriously more likely to have the outcome of interest. For instance, clinical protocols might lead to a dataset in which patients with pacemakers are disproportionately likely to have congestive heart failure. This skew can lead to models that take shortcuts by heavily relying on the biased attribute. We explore this problem across a number of attributes in the context of diagnosing the cause of acute hypoxemic respiratory failure. Applied to chest X-rays, we show that i) deep nets can accurately identify many patient attributes including sex (AUROC = 0.96) and age (AUROC ≥ 0.90), ii) they tend to exploit correlations between such attributes and the outcome label when learning to predict a diagnosis, leading to poor performance when such correlations do not hold in the test population (e.g., everyone in the test set is male), and iii) a simple transfer learning approach is surprisingly effective at preventing the shortcut and promoting good generalization performance. On the task of diagnosing congestive heart failure based on a set of chest X-rays skewed towards older patients (age ≥ 63), the proposed approach improves generalization over standard training from 0.66 (95% CI: 0.54-0.77) to 0.84 (95% CI: 0.73-0.92) AUROC. While simple, the proposed approach has the potential to improve the performance of models across populations by encouraging reliance on clinically relevant manifestations of disease, i.e., those that a clinician would use to make a diagnosis. Copyright © 2020, The Authors. All rights reserved.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

A systems biology approach to identifying genetic factors affected by aging, lifestyle factors, and type 2 diabetes that influences Parkinson's disease progression

引用

Informatics in Medicine Unlocked 2020年 21卷

作者： Sakib, Najmus Chowdhury, Utpala Nanda Islam, M. Babul Ahmad, Shamim Moni, Mohammad Ali Department of Electrical and Electronic Engineering University of Rajshahi Rajshahi Bangladesh Department of Computer Science and Engineering University of Rajshahi Rajshahi Bangladesh Pabna University of Science and Technology Pabna Bangladesh WHO Collaborating Centre on eHealth School of Public Health and Community Medicine Faculty of Medicine UNSW Sydney Australia

The underlying causes of Parkinson's disease (PD) remain unclear, making it difficult to determine how particular aspects of a patient's health may affect PD development and their treatment response. One alternative approach of investigating interactions between PD and co-morbidities (or other health factors) is to compare the dysregulated cellular or gene pathways they share with affected tissues. To achieve this, we developed a quantitative framework to reveal significant interactions between PD and other significant health factors that may be impacted. In this study, we analyzed gene expression microarray data from tissues affected by PD, type II diabetes (T2D), aging (AG), sedentary lifestyle (SL), high dietary fat (HFD), hypercholesterolemia (HC), high body fat (HBF), high red meat diet (RMD), high alcohol consumption (AC), smoking (SM) and control datasets. We have developed genetic associations of these factors with PD, based on the neighborhood-based benchmarking and multilayer network topology. We identified 1323 differentially expressed genes (DEGs) in the PD patients compared to healthy controls, 779 genes with down-regulated expression and 544 genes up-regulated. 69 dysregulated genes were common to PD and AC datasets;PD datasets also shared, respectively, 51, 45, 43 and 42 DEGs with the T2D, AG, HFD and HBF datasets. Ontological and pathway analyses identified significant gene ontology and molecular pathways with the potential to enhance our understanding of the fundamental molecular factors associated with PD progression. Moreover, we employed a validation gene expression dataset along with Mendelian Inheritance in Man (OMIM) and dbGaP as gold benchmark databases to validate the identified PD associated genes and molecular pathways. Our formulated methodologies used a network-based approach and identified a number of significant genes and pathways that may particularly affect PD, signifying how PD incidence and development may be influenced by the risk

关键词： Aging Alcohol consumption Lifestyle Network-based Parkinson's disease Smoking Topological Type II diabetes

来源：评论

学校读者我要写书评

暂无评论

Three Dimensional Light Microscopy: Imaging &Corrections for Quantitative Analysis

引用

Microscopy and Microanalysis 2020年第S2期5卷 522-523页

作者： J.N. Turner W. hain D.H. Szarowski S. Lasek L. Kam A. Can K. Al-Kofahi B. Roysam Wadsworth Center New York State Department of Health The University at Albany Albany New York12201 School of Public Health The University at Albany Albany New York12201 Biomedical Engineering Rensselaer Polytechnic Institute Troy New York12180 Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy New York12180

There are several forms of three-dimensional (3-D) light microscopy but all utilize the principle of optical section recording, i.e. the 3-D image is a sequence of two-dimensional (2-D) images called optical sections. The optical sections are particular focal planes formed within the thick specimen and usually correspond to the conventional image projections recorded in a light microscope, referred to as x,y projections. The optical sections are recorded for a sequence of focus- or z-positions. This “stack” of 2-D images is the data set for the 3-D image. If quantitative analysis is to be performed on the 3-D images, the choice of the z-dimension increment between 2-D images is especially important, and its value may be more or less critical depending on the analysis algorithm used. A reasonable starting value for this dimension is the depth-of-field of the objective lens, but the actual value may have to be smaller to optimize the image analysis or larger to decrease the influence of photobleaching. The most photostable dyes should be selected and the specimen should be mounted in index-of-refraction matching media with an *** image resolution in all three-dimensions is determined by the 3-D point-spread-function (psf), and as a rough rule of thumb the z-resolution is degraded by a factor of 3 relative to the x,y resolution. To achieve or at least approach isotropic resolution the 3-D image can be deconvolved. Figure 1 shows the 2-D maximum value projection of a 3-D image of a cultured glial cell dual labeled for actin and vinculin before and after deconvolution. The actin fibers and vinculin focal contacts are more clearly resolved after deconvolution. Although a single cultured cell might traditionally be considered a thin object, it is really a thick object if the desired spatial resolution is less than the thickness of the cell. It is desirable to image as deep into a thick object as possible to maximize the tissue volume sampled. However, it has be

关键词：

来源：评论

学校读者我要写书评

暂无评论

Correction: Machine learning-based prognostic model for 30-day mortality prediction in Sepsis-3

引用

BMC medical informatics and decision making 2024年第1期24卷 264页

作者： Md Sohanur Rahman Khandaker Reajul Islam Johayra Prithula Jaya Kumar Mufti Mahmud Mohammed Fasihul Alam Mamun Bin Ibne Reaz Abdulrahman Alqahtani Muhammad E H Chowdhury Department of Electrical and Electronics Engineering University of Dhaka Dhaka 1000 Bangladesh. Department of Physiology Faculty of Medicine University Kebangsaan Malaysia 56000 Kuala Lumpur Malaysia. Department of Physiology Faculty of Medicine University Kebangsaan Malaysia 56000 Kuala Lumpur Malaysia. jayakumar@ukm.edu.my. Department of Computer Science Nottingham Trent University Clifton Lane Nottingham NG11 8NS UK. Department of Public Health College of Health Sciences QU Health Qatar University Doha 2713 Qatar. Department of Electrical Engineering Independent University Bangladesh Dhaka Bangladesh. Department of Biomedical Technology College of Applied Medical Sciences in Al-Kharj Prince Sattam Bin Abdulaziz University Al-Kharj 11942 Saudi Arabia. Department of Medical Equipment Technology College of Applied Medical Science Majmaah University Majmaah City 11952 Saudi Arabia. Department of Electrical Engineering Qatar University Doha 2713 Qatar. mchowdhury@qu.edu.qa.

来源：评论

学校读者我要写书评

暂无评论

The false hope of current approaches to explainable artificial intelligence in health care

引用

The Lancet Digital Health 2021年第11期3卷 e745-e750页

作者： Ghassemi, Marzyeh Oakden-Rayner, Luke Beam, Andrew L Department of Electrical Engineering and Computer Science and Institute for Medical and Evaluative Sciences Massachusetts Institute of Technology Cambridge MA United States Vector Institute Toronto ON Canada Australian Institute for Machine Learning University of Adelaide Adelaide SA Australia CAUSALab and Department of Epidemiology Harvard T H Chan School of Public Health Boston MA United States Department of Biomedical Informatics Harvard Medical School Boston MA United States

The black-box nature of current artificial intelligence (AI) has caused some to question whether AI must be explainable to be used in high-stakes scenarios such as medicine. It has been argued that explainable AI will engender trust with the health-care workforce, provide transparency into the AI decision making process, and potentially mitigate various kinds of bias. In this Viewpoint, we argue that this argument represents a false hope for explainable AI and that current explainability methods are unlikely to achieve these goals for patient-level decision support. We provide an overview of current explainability techniques and highlight how various failure cases can cause problems for decision making for individual patients. In the absence of suitable explainability methods, we advocate for rigorous internal and external validation of AI models as a more direct means of achieving the goals often associated with explainability, and we caution against having explainability be a requirement for clinically deployed models. © 2021 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep-learning-enabled Brain Hemodynamic Mapping Using Resting-state fMRI

arXiv

引用

arXiv 2022年

作者： Hou, Xirui Guo, Pengfei Wang, Puyang Liu, Peiying Lin, Doris D.M. Fan, Hongli Li, Yang Wei, Zhiliang Lin, Zixuan Jiang, Dengrong Jin, Jin Kelly, Catherine Pillai, Jay J. Huang, Judy Pinho, Marco C. Thomas, Binu P. Welch, Babu G. Park, Denise C. Patel, Vishal M. Hillis, Argye E. Lu, Hanzhang Department of Biomedical Engineering Johns Hopkins University School of Medicine 360 Park Building 600 N. Wolfe Street BaltimoreMD21287 United States The Russell H. Morgan Department of Radiology and Radiological Science Johns Hopkins University School of Medicine BaltimoreMD United States Department of Computer Science Johns Hopkins University BaltimoreMD United States Department of Electrical and Computer Engineering Johns Hopkins University BaltimoreMD United States Department of Diagnostic Radiology and Nuclear Medicine University of Maryland School of Medicine BaltimoreMD United States F.M. Kirby Research Center for Functional Brain Imaging Kennedy Krieger Institute BaltimoreMD United States Department of Biostatistics Johns Hopkins Bloomberg School of Public Health BaltimoreMD United States Department of Neurology Johns Hopkins University School of Medicine BaltimoreMD United States Department of Neurosurgery Johns Hopkins University School of Medicine BaltimoreMD United States Department of Radiology UT Southwestern Medical Center DallasTX United States Department of Neurologic Surgery UT Southwestern Medical Center DallasTX United States Center for Vital Longevity School of Behavioral and Brain Sciences University of Texas at Dallas DallasTX United States

Cerebrovascular disease is a leading cause of death globally. Prevention and early intervention are known to be the most effective forms of its management. Non-invasive imaging methods hold great promises for early stratification, but at present lack the sensitivity for personalized prognosis. Resting-state functional magnetic resonance imaging (rs-fMRI), a powerful tool previously used for mapping neural activity, is available in most hospitals. Here we show that rs-fMRI can be used to map cerebral hemodynamic function and delineate impairment. By exploiting time variations in breathing pattern during rs-fMRI, deep learning enables reproducible mapping of cerebrovascular reactivity (CVR) and bolus arrive time (BAT) of the human brain using resting-state CO2fluctuations as a natural "contrast media". The deep-learning network was trained with CVR and BAT maps obtained with a reference method of CO2-inhalation MRI, which included data from young and older healthy subjects and patients with Moyamoya disease and brain tumors. We demonstrate the performance of deep-learning cerebrovascular mapping in the detection of vascular abnormalities, evaluation of revascularization effects, and vascular alterations in normal aging. In addition, cerebrovascular maps obtained with the proposed method exhibited excellent reproducibility in both healthy volunteers and stroke patients. Deep-learning resting-state vascular imaging has the potential to become a useful tool in clinical cerebrovascular imaging. Copyright © 2022, The Authors. All rights reserved.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

The medical algorithmic audit (vol 4, pg e384, 2022)

引用

LANCET DIGITAL HEALTH 2022年第6期4卷 E405-E405页

作者： Liu, X. Glocker, B. McCradden, M. M. Ghassemi, M. Denniston, A. K. Oakden-Rayner, L. Academic Unit of Ophthalmology Institute of Inflammation and Ageing College of Medical and Dental Sciences University of Birmingham UK Department of Ophthalmology University Hospitals Birmingham NHS Foundation Trust Birmingham UK Moorfields Eye Hospital NHS Foundation Trust London UK Health Data Research UK London UK Birmingham Health Partners Centre for Regulatory Science and Innovation University of Birmingham Birmingham UK Biomedical Image Analysis Group Department of Computing Imperial College London London UK The Hospital for Sick Children Toronto ON Canada Dalla Lana School of Public Health Toronto ON Canada Institute for Medical Engineering and Science and Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology Cambridge MA USA National Institute of Health Research Biomedical Research Centre for Ophthalmology Moorfields Hospital London NHS Foundation Trust London UK University College London Institute of Ophthalmology London UK Australian Institute for Machine Learning University of Adelaide Adelaide SA Australia. lauren.oakden-rayner@adelaide.edu.au

Artificial intelligence systems for health care, like any other medical device, have the potential to fail. However, specific qualities of artificial intelligence systems, such as the tendency to learn spurious correlates in training data, poor generalisability to new deployment settings, and a paucity of reliable explainability mechanisms, mean they can yield unpredictable errors that might be entirely missed without proactive investigation. We propose a medical algorithmic audit framework that guides the auditor through a process of considering potential algorithmic errors in the context of a clinical task, mapping the components that might contribute to the occurrence of errors, and anticipating their potential consequences. We suggest several approaches for testing algorithmic errors, including exploratory error analysis, subgroup testing, and adversarial testing, and provide examples from our own work and previous studies. The medical algorithmic audit is a tool that can be used to better understand the weaknesses of an artificial intelligence system and put in place mechanisms to mitigate their impact. We propose that safety monitoring and medical algorithmic auditing should be a joint responsibility between users and developers, and encourage the use of feedback mechanisms between these groups to promote learning and maintain safe deployment of artificial intelligence *** © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license. Published by Elsevier Ltd.. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Gradient Boosted Tree Based Feature Selection and Parkinson’s Disease Classification

Gradient Boosted Tree Based Feature Selection and Parkinson...

引用

International Conference on Science and Technology-computer (ICST)

作者： Adi Wijaya Kharis Wini Prastuti Department of Electrical and Information Engineering Universitas Gadjah Mada Yogyakarta Indonesia Department of Public Policy Management Universitas Gadjah Mada Yogyakarta Indonesia

ISBN: (数字)9781728123691

ISBN: (纸本)9781728123707

Parkinson's Disease (PD) is one of the most common neurodegenerative disorder with increasing prevalence in the world. Many studies have been conducted to detect or diagnose PD. Recently, voice signal based PD detection has been conducted rapidly. However, the performance of the detection still not excellent, hence the need for excellent detection still an open issue in PD detection. In this study, PD detection based on voice signal was conducted using public dataset. Since the dataset consist of many features, the proposed method combines feature selection mechanism and classification based on Gradient Boosted Tree (GBT) algorithm. Feature selection was conducted by using feature weighting from GBT. The experimental results show that the proposed method outperforms prior research with the accuracy about 0.92 with the minimum number of features. This result proves that GBT able to perform feature selection and good classifier. This concludes that the proposed method is promising as PD detection based on voice signal analysis.

关键词： Feature extraction Support vector machines Parkinson's disease Boosting Classification algorithms Signal analysis

来源：评论

学校读者我要写书评

暂无评论

The cysteine-altering p.R544C variant in the NOTCH3 gene is a probable candidate for blood pressure and relevant traits in the Taiwan Biobank

引用

Journal of neurology 2023年第11期270卷 5536-5544页

作者： Eugene Lin Po-Hsiu Kuo Yu-Li Liu Tso-Jen Wang Albert C Yang Shih-Jen Tsai Department of Biostatistics University of Washington 3980 15th Avenue NE Box 351617 Seattle WA 98195 USA. lines@uw.edu. Department of Electrical and Computer Engineering University of Washington Seattle WA 98195 USA. lines@uw.edu. Graduate Institute of Biomedical Sciences China Medical University Taichung 40402 Taiwan. lines@uw.edu. Department of Public Health Institute of Epidemiology and Preventive Medicine National Taiwan University Taipei 10617 Taiwan. Department of Psychiatry National Taiwan University Hospital Taipei 100 Taiwan. Center for Neuropsychiatric Research National Health Research Institutes Miaoli County 35053 Taiwan. Taoyuan Psychiatric Center Ministry of Health and Welfare Taoyuan 33058 Taiwan. Brain Research Center National Yang Ming Chiao Tung University Taipei 11221 Taiwan. Department of Psychiatry Taipei Veterans General Hospital No. 201 Shih-Pai Road Sec. 2 Taipei 11217 Taiwan. tsai610913@***. Division of Psychiatry National Yang Ming Chiao Tung University Taipei 112304 Taiwan. tsai610913@***.

BACKGROUND:The cysteine-altering variants in NOTCH3 have been suggested to be associated with stroke, dementia, and cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), where aberrant blood pressure levels represent the characteristics of these diseases. We aimed to assess whether the cysteine-altering ***544Cys (p.R544C; rs201118034) variant and common single nucleotide variants (SNVs) in NOTCH3 could contribute to systolic and diastolic blood pressure and related phenotypes in the Taiwan Biobank. METHODS:We employed a discovery sample of 68,925 individuals, an independent replication sample of 45,676 individuals, and a combined/total sample of 114,601 individuals; all from the Taiwan Biobank. Blood pressure, such as systolic and diastolic blood pressure, was measured for all participants. Association was evaluated using a general linear model, where results were considered statistically significant if the P value < 0.05 divided by the number of independent tests per model. RESULTS:From our analysis, we identified and replicated three novel candidates for blood pressure that have not previously been reported: the cysteine-altering p.R544C variant for systolic blood pressure, the common SNV rs11669950 for diastolic blood pressure, and the common SNV rs4808235 for diastolic blood pressure. We also generalized two previously identified SNVs (i.e., rs10418305 and rs7408868) in NOTCH3 for blood pressure in European and non-Taiwanese East Asian populations to the Taiwanese population. Moreover, the participants with NOTCH3 p.R544C had an increased stroke frequency (P < 1.0 × 10) and a higher dementia frequency (P = 2.0 × 10) compared with the whole Taiwan Biobank population in the combined/total sample. CONCLUSION:NOTCH3 is a strong candidate for a role in stroke, dementia, and CADASIL, which has previously been linked to blood pressure changes. While our preliminary study suggests that NOTCH3 p.R544C may influence blood

关键词： Biobank Blood pressure CADASIL Dementia NOTCH3 Neurology Stroke

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：