Automatic Video Annotation of Human Health Care Action via Clustering

2020 ◽  
Vol 10 (10) ◽  
pp. 2512-2521
Keyword(s):  
Health Care ◽  
Human Health ◽  
Large Scale ◽  
Image Annotation ◽  
Spinal Injury ◽  
Health Sector ◽  
Video Annotation ◽  
Adjusted Rand Index ◽  
Crowd Sourcing ◽  
Data Set

Vision-based activity monitoring provides applications that revolutionized the e-health sector. Considering the potential of crowdsourcing data, to develop large scale applications, the researchers are working on consolidating smart hospital with crowd sourcing data. For creating a meaningful pattern from such huge data, a key challenge is that it needs to be annotated. Especially, the annotation of medical images plays an important role in providing pervasive health services. Although, multiple image annotation methods such as manual and semi-supervised exist. However, high cost and computation time remains a major issue. To overcome the abovementioned issues, a methodology is proposed for automatic annotation of images. The proposed approach is based on three tires namely frame extraction, interest point's generation, and clustering. Since the medical imaging lacks an appropriate dataset for our experimentation. Consequently, we have introduced a new dataset of Human Health care Actions (HHA). The data set comprises of videos related to multiple medical emergencies such as allergy reactions, burn, asthma, brain injury, bleeding, poisoning, heart attack, choking and spinal injury. We have also proposed an evaluation model to assess the effectiveness of the proposed methodology. The promising results of the proposed technique indicate the effectiveness of 78% in terms of Adjusted Rand Index. Furthermore, to investigate the effectiveness of the proposed technique, a comparison is made, by training the neural network classifier with annotated labels generated by proposed methodology and other existing techniques such as semi-supervised and manual methods. The overall precision of the proposed methodology is 0.75 (i.e., 75%) and semi-supervised learning is 0.69 (69%).

Semantic Fusion of Image Annotation

2011 ◽  
Vol 268-270 ◽  
pp. 1386-1389
Author(s):  
Xiao Ying Wu ◽  
Yun Juan Liang ◽  
Li Li ◽  
Li Juan Ma
Keyword(s):  
Large Scale ◽  
Image Annotation ◽  
Semantic Annotation ◽  
Probability Model ◽  
Joint Probability ◽  
Training Data ◽  
Data Set ◽  
Other Information ◽  
Image Area ◽  
Representative Image

In this paper, improve the image annotation with semantic meaning, and name the new algorithm for semantic fusion of image annotation, that is a image is given to be labeled, use of training data set, the word set, and a collection of image area and other information to establish the probability model ,estimates the joint probability by word and given image areas.The probability value as the size, combined with keywords relevant table that integrates lexical semantics to extract keywords as the most representative image semantic annotation results. The algorithm can effectively use large-scale training data with rich annotation, so as to achieve better recall and precision than the existing automatic image annotation ,and validate the algorithm in the Corel data set.

Public–Private Partnerships in the health sector: the Danish experience

2008 ◽  
Vol 3 (2) ◽  
pp. 141-163 ◽  
Author(s):  
KARSTEN VRANGBÆK*
Keyword(s):  
Health Care ◽  
Large Scale ◽  
Health Sector ◽  
Theoretical Models ◽  
Public Private Partnerships ◽  
Public And Private ◽  
Theoretical Approaches ◽  
General Terms ◽  
Danish Health ◽  
Private And Public

AbstractThis article investigates the current use of Public–Private Partnerships (PPP) in the Danish health sector based on an initial discussion of theoretical approaches that analyze PPP. The empirical analysis concludes that PPP has been used very sparsely in the Danish health sector. There are few examples of large-scale partnership projects with joint investment and risk taking, but a number of smaller partnerships such as jointly owned companies at the regional level. When defining PPP more broadly, we can identify a long tradition for various types of collaboration between public and private actors in health care in Denmark. An analysis of the regulatory environment is offered as an explanation for the limited use of PPPs in Denmark. Major political and institutional actors at the central level differ in their enthusiasm for the PPP concept, and the regulatory framework is somewhat uncertain. A number of general issues and concerns related to PPPs are also discussed. It is suggested that a risk-based framework can be useful for mapping the potential and challenges for both private and public partners. Such a framework can be used to feed into game theoretical models of pros and cons for PPP projects. In general terms, it is concluded that more empirical research is needed for the assessment of the various risk factors involved in using PPPs in health care. Most PPPs are still very young, and the evidence on performance and broader governance issues is only just emerging. Ideally, such assessments should include comparisons with a purely public alternative.

scholarly journals Comparative Analysis of Supervised and Unsupervised Approaches Applied to Large-Scale “In The Wild” Face Verification

Symmetry ◽  
2020 ◽  
Vol 12 (11) ◽  
pp. 1832
Author(s):  
Tomasz Hachaj ◽  
Patryk Mazurek
Keyword(s):  
Pattern Recognition ◽  
Large Scale ◽  
Statistical Significance ◽  
Ground Truth ◽  
Classification Algorithm ◽  
Adjusted Rand Index ◽  
Face Verification ◽  
Data Set ◽  
In The Wild ◽  
Unsupervised Approaches

Deep learning-based feature extraction methods and transfer learning have become common approaches in the field of pattern recognition. Deep convolutional neural networks trained using tripled-based loss functions allow for the generation of face embeddings, which can be directly applied to face verification and clustering. Knowledge about the ground truth of face identities might improve the effectiveness of the final classification algorithm; however, it is also possible to use ground truth clusters previously discovered using an unsupervised approach. The aim of this paper is to evaluate the potential improvement of classification results of state-of-the-art supervised classification methods trained with and without ground truth knowledge. In this study, we use two sufficiently large data sets containing more than 200,000 “taken in the wild” images, each with various resolutions, visual quality, and face poses which, in our opinion, guarantee the statistical significance of the results. We examine several clustering and supervised pattern recognition algorithms and find that knowledge about the ground truth has a very small influence on the Fowlkes–Mallows score (FMS) of the classification algorithm. In the case of the classification algorithm that obtained the highest accuracy in our experiment, the FMS improved by only 5.3% (from 0.749 to 0.791) in the first data set and by 6.6% (from 0.652 to 0.718) in the second data set. Our results show that, beside highly secure systems in which face verification is a key component, face identities discovered by unsupervised approaches can be safely used for training supervised classifiers. We also found that the Silhouette Coefficient (SC) of unsupervised clustering is positively correlated with the Adjusted Rand Index, V-measure score, and Fowlkes–Mallows score and, so, we can use the SC as an indicator of clustering performance when the ground truth of face identities is not known. All of these conclusions are important findings for large-scale face verification problems. The reason for this is the fact that skipping the verification of people’s identities before supervised training saves a lot of time and resources.

Image/Video Semantic Analysis by Semi-Supervised Learning

2011 ◽  
pp. 183-210
Author(s):  
Jinhui Tang ◽  
Xian-Sheng Hua ◽  
Meng Wang
Keyword(s):  
Supervised Learning ◽  
Large Scale ◽  
Semantic Analysis ◽  
Image Annotation ◽  
Structural Difference ◽  
Video Annotation ◽  
Learning Methods ◽  
Manifold Ranking ◽  
Training Samples ◽  
Structure Sensitive

The insufficiency of labeled training samples is a major obstacle in automatic semantic analysis of large scale image/video database. Semi-supervised learning, which attempts to learn from both labeled and unlabeled data, is a promising approach to tackle this problem. As a major family of semi-supervised learning, graph-based methods have attracted more and more recent research. In this chapter, a brief introduction is given on popular semi-supervised learning methods, especially the graph-based methods, as well as their applications in the area of image annotation, video annotation, and image retrieval. It is well known that the pair-wise similarity is an essential factor in graph propagation based semisupervised learning methods. A novel graph-based semi-supervised learning method, named Structure- Sensitive Anisotropic Manifold Ranking (SSAniMR), is derived from a PDE based anisotropic diffusion framework. Instead of using Euclidean distance only, SSAniMR further takes local structural difference into account to more accurately measure pair-wise similarity. Finally some future directions of using semi-supervised learning to analyze the multimedia content are discussed.

scholarly journals Adapting research to the 21st century – the Swedish Biobank Register

2012 ◽  
Vol 21 (2) ◽  
Author(s):  
L. Norlin ◽  
M. Fransson ◽  
S. Eaker ◽  
G. Elinder ◽  
J.-E. Litton
Keyword(s):  
Health Care ◽  
Health Care Quality ◽  
Statistical Power ◽  
Large Scale ◽  
Science Research ◽  
Population Based ◽  
Care Quality ◽  
Data Set ◽  
Large Patient ◽  
Medical Health Care

<p>In Sweden, there are currently nearly 600 biobanks. The Swedish Biobank Register (SBR) is an on-going national investment by the county councils working to capture information in one database about all biobank samples collected from patients attending the Swedish medical health care. The aim of the SBR is to gather enough information about biobank samples to be able to physically trace the samples.</p><p>The BioBanking and Molecular Resource Infrastructure of Sweden (BBMRI.se) has been given the task of extending the SBR Information System (IS) with functionality useful for research in connection to health care, quality registers and large patient cohorts. The research extension is called BBMRI catalogue over sample collections for research. To achieve this, the SBR-IS will be extended with attributes useful for both epidemiological and clinical research enabling authorized researchers to search for samples stored at non-clinical biobanks nationwide. The Swedish Biobank Register, together with the BBMRI research catalogue, will be a unique resource for research. SBR-IS will contain information about biobank samples collected by both clinical and population-based biobanks specifically established for research purposes but BBMRI.se researchers will only be granted access to data related to population-based biobanks. As BBMRI.se is the Swedish hub of the pan-European biobank project BBMRI, whose goal is to promote excellence and efficacy in European life science research, the BBMRI research catalogue will also be made compatible with the European register by applying its minimum data set describing biobanks and their objects. In this paper we describe the implementation. Our belief is that it will pave the way for connecting biobanks on an international level as well as stimulate collaborations and optimize usage of biobank samples. In the long run, patients and sample donors will benefit as new results with high statistical power emerge from large scale studies.</p>

scholarly journals ORGANO-PHOSPHONATE, ORGANO-HALOGENATTED, CYANIDES AND ARSENIC TOXIC IMPACT ON THE HUMAN HEALTH CARE CONDITION

2019 ◽  
Vol 5 (15) ◽  
pp. 1357-1362
Author(s):  
Venelin Terziev ◽  
Stoyanka Petkova - Georgieva
Keyword(s):  
Health Care ◽  
Human Health ◽  
Large Scale ◽  
Deep Understanding ◽  
Chemical Compounds ◽  
Chemical Weapon ◽  
General Description ◽  
Industrial Chemicals ◽  
Toxic Industrial Chemicals ◽  
Toxic Impact

This research describes the toxic impact on the health condition of living nature because of the inevitable use of toxic chemical compounds (TCC) and toxic industrial chemicals (TICs). To reach this aim, the following objectives were established: to define the large-scale industrial chemicals from the Schedules of the Chemical weapon convention; to investigate the role of each chemical in production process; to determine the current amounts of substances used in the process and find the alternatives for their replacement. The research was conducted on the base of only open-source literature. The general description of the chemicals use was found in electronic and paper encyclopaedias, more deep understanding – received after study of scientific and technological periodics and patents from electronic bases. The content in this review is focused on the toxic impact of the organo – phosphonate, organo – halogenated, cyanides and arsenic toxic chemicals compounds. Keywords: impact, human health care, toxis chemical compounds, toxic industial chemicals.

scholarly journals Current Trends and Future Directions of Large Scale Image and Video Annotation: Observations From Four Years of BIIGLE 2.0

2021 ◽  
Vol 8 ◽  
Author(s):  
Martin Zurowietz ◽  
Tim W. Nattkemper
Keyword(s):  
Machine Learning ◽  
Large Scale ◽  
Image Annotation ◽  
Training Data ◽  
Video Annotation ◽  
Computer Assisted ◽  
Marine Science ◽  
Imaging Data ◽  
Video Frames ◽  
Increasing Demand

Marine imaging has evolved from small, narrowly focussed applications to large-scale applications covering areas of several hundred square kilometers or time series covering observation periods of several months. The analysis and interpretation of the accumulating large volume of digital images or videos will continue to challenge the marine science community to keep this process efficient and effective. It is safe to say that any strategy will rely on some software platform supporting manual image and video annotation, either for a direct manual annotation-based analysis or for collecting training data to deploy a machine learning–based approach for (semi-)automatic annotation. This paper describes how computer-assisted manual full-frame image and video annotation is currently performed in marine science and how it can evolve to keep up with the increasing demand for image and video annotation and the growing volume of imaging data. As an example, observations are presented how the image and video annotation tool BIIGLE 2.0 has been used by an international community of more than one thousand users in the last 4 years. In addition, new features and tools are presented to show how BIIGLE 2.0 has evolved over the same time period: video annotation, support for large images in the gigapixel range, machine learning assisted image annotation, improved mobility and affordability, application instance federation and enhanced label tree collaboration. The observations indicate that, despite novel concepts and tools introduced by BIIGLE 2.0, full-frame image and video annotation is still mostly done in the same way as two decades ago, where single users annotated subsets of image collections or single video frames with limited computational support. We encourage researchers to review their protocols for education and annotation, making use of newer technologies and tools to improve the efficiency and effectivity of image and video annotation in marine science.

scholarly journals SPRING POINT DETECTION OF HIGH RESOLUTION IMAGE BASED ON YOLOV3

2020 ◽  
Vol XLIII-B2-2020 ◽  
pp. 645-650
Author(s):  
J. Wu ◽  
Z. Zhang ◽  
G. Huang ◽  
G. Ma
Keyword(s):  
Remote Sensing ◽  
Large Scale ◽  
Image Annotation ◽  
Remote Sensing Image ◽  
Acquisition Time ◽  
Data Set ◽  
Sensing Technology ◽  
Belt And Road ◽  
Point Detection ◽  
The Belt And Road

Abstract. The Xinjiang region of China is a vast and sparsely populated area with complex topography, surrounded by basins and mountains, and its geomorphological features and water circulation process make the traditional spring water resource acquisition time-consuming, labor-consuming and inaccurate. Remote Sensing Technology has the advantages of large scale, periodicity, timeliness and comprehensiveness in target detection. In order to realize the artificial intelligence detection of springs in Xinjiang, this paper presents a method of detecting springs in remote sensing image based on the YOLOV3 network framework, based on the data set of 512 * 512 by using 0.8m remote sensing image annotation, a model of recognition of spring point based on Yolov3 network is constructed and trained. The results show that the map of spring point is 0.973, which is the basis of monitoring and protecting the natural environment in the Belt and Road Initiatives.

The effect of lean on performance: a longitudinal study of US Hospitals

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Huilan Zhang
Keyword(s):  
Health Care ◽  
Least Squares ◽  
Large Scale ◽  
Probit Model ◽  
Ordinary Least Squares ◽  
Hospital Performance ◽  
Charge Ratio ◽  
Data Set ◽  
Lean Implementation ◽  
Content Type

Purpose There has been little empirical research focused on the effect of lean on hospital performance in the form of a consolidated methodology. This paper aims to apply a more sophisticated approach to examine whether hospitals’ decision for lean implementation is endogenous and test the effects of lean on hospital performance. Design/methodology/approach This study uses a publicly available data set of hospitals across the USA from 2002 to 2019 and performs two-stage least squares (2SLS) analysis. In the first stage, a probit model is used to estimate hospitals’ decision to implement lean. The fitted probability values from the first stage are used in the second stage to test the relationship between lean and hospital performance. Ordinary least squares (OLS) regression results are compared with those of the 2SLS approach. Findings The decision to implement lean is significantly associated with hospital-specific characteristics (the complexity of care, size and cost-to-charge ratio), indicating hospitals’ decision for lean implementation is endogenous. Moreover, there is strong evidence that lean implementation is positively associated with hospital financial and operational performance. The Hausman F-tests confirm the presence of endogeneity and this, in turn, suggests that OLS regressions result in unreliable estimates. Practical implications The findings of this study can help hospital managers benchmark performance and explore opportunities for profit and efficiency improvement. The findings are also relevant to policymakers who strive to lower health-care spending. Originality/value This study is motivated by the challenges facing the health-care industry. This study is among the first to investigate endogeneity in lean implementation and the association between lean and hospital performance using large-scale archival panel data. The use of the 2SLS approach provides more confidence in statistical findings.

scholarly journals A Privacy Preserving Upload Model for Crowd Sourced Health Care Monitoring System

2020 ◽  
Vol 9 (3) ◽  
pp. 3337-3346
Keyword(s):  
Health Care ◽  
Quality Of Health Care ◽  
Large Scale ◽  
Third Party ◽  
Crowd Sourcing ◽  
Privacy Requirements ◽  
Health Care Data ◽  
Firm Commitment ◽  
Health Care Monitoring

In recent years governments become more concern about health care monitoring which cannot be accomplished without enhancing trust on underlying system as various citizens hesitate to upload their sample because of privacy reasons and obviously the governmental decisions are based on the data collected by various PHCs and third party medical agencies. The accuracy and authenticity of this third party owned data is always doubtful. Crowd sourcing(a collaborative framework) make its sound presence in development of large scale health projects Scientist also impressed from crowd sourcing which is a faster and better alternative to traditional methods for predicting and monitoring infectious diseases. However the success of this type of crowd sourcing depends on the trust on underlying system as the user is always looking firm commitment to preserve their privacy and win a promise of not being re-identified later. Here in this work we suggest a privacy protecting framework for upload process which could fulfill user's diverse privacy requirements while guaranteeing the quality of health care data.

Sign in / Sign up

Export Citation Format

Share Document