- April 8, 2020
Interactive Data Visualization
Theaim of the research is to come up with tools for interactivevisualization of HIV group data. An HIV data visualization tool wasdeveloped by using R-statistical Lan-gauge. The tool requires thatdata structure follows the HIV data exchange protocol. Besides, thetool requires that the implementation conforms to the CCASA net data.The interactive visualization of HIV cohort tool presents data by useof three classes of plots. They include longitudinal, heat maps andbubble plots. All these plots are presented using CCASA net datainvestigation trends. The group is aimed at expanding the researchprocess by coming up with additional shareable tools. Additionally,the network hopes that it will further open data collection by theHIV research community.
Datavisualization is very crucial. Researchers have found out that datavisualization explores analysis of data structures it interpretstrends of events in a population over a given period. Also, datavisualization communicates the inferences that have been obtainedfrom the data that has been analyzed. Data visualization helps inunderstanding HIV disease (Carroll, 2014). Besides, it has been foundout that, the future developers are going to rely on the broadercontext of the data that is available. The developers will as welllook at team collaboration and interdisciplinary needs. The existingtools have limited learning curves. They attract the users they arefree and also transparent.
Datavisualization is represented by the use of three curves. The firstcurve is a longitudinal plot. The plot is motivated by commongraphics such as spaghetti, density and Kaplan-Meier curves. Thecurves are used to describe HIV AIDS outcome in a given level orcountry (Robinson, 2011). The density curve is a graph that is usedto show the probability of events. The longitudinal and the smoothcurves are viewed in line with the event probability curves at thesame time. The two are used to inspect outcomes grouped according toclasses of HIV patients. The outcome includes the mortality rate andbirth rate. Besides the longitudinal plots, there are the bubblecurves. The bubble curves show variation in indicators over a givenperiod. They give time for observation of group level dynamics. Theyare three-dimensional plots. They have x, y and z planes.
Thethird data visualization curve is the heat map. The heat map exhibitsborders of countries that are filled in with dark colors for higherproportion and light colors for low proportions. They are used toshow the burden of HIV epidemic worldwide. A cd4 sample R script isused to show how an individual can generate dataset using patientlevel data. The CD4 is a lab test that is used to measure the CD4 Tlymphocytes of an individual’s blood (Luz, 2015).
Ithas been observed that the density curves are used to measure therecent distribution of the CD4 count. Additionally, the Kaplan-Meiercurves are used to estimate the possibility of death. According toresearch, it is evident that patients initiating cART instantlyseparate into lower and higher CD4 status. It is important to notethat the conditional survival of the two groups after one yearbecomes similar. The results are usually obtained after CD4 cellcounts.
Thetop panel of the bubble plots shows regions that correspond to theobserved proportions. They also show bubble size that has the sameproportion to that of newly found individuals. The bottom panelindicates marginal allocations. The plot shows bubble proportionrepresenting each region as time progresses. From the curves, it isobserved that most sites show that the proportion of newly enrolledindividuals low CD4 count decrease with time. The bottom panel showsa marginal representation of patients with HIV diagnosis (Meridith,2016). Besides, it is evident that the bottom panel shows low CD4count that represents a non-zero share of a country level enrollmentpopulation. The results represent a situation whereby the objectiveand subjective measures fails to agree with the data available.
Themajor aim of the CCASAnet is to aid the HIV researchers to create alarger data visualization tool that will enable them to have theinsight of the HIV outcomes. The longitudinal curves are used for toshow the changes in the CD4 count they are used to measure thepossibility of AIDS-defining events. Additionally, the survivalcurves are used to measure hemoglobin in the body and the HIV viralrod. The bubble plots are used to show the movement of the mainindicators across a given group of individuals over a given period.The heat maps are used to show a spatial element of group data aboutthe population trends. The learning curve can be flattened by use ofgraphical user interface (GUI). The GUI allows one to put datadirectly without manipulation, unlike the CSV document that requiresspecifications to be edited. The coma separated values (CSV) documentis a file format that is used to store data in tables such as aspreadsheet.
Ina nutshell, further use of the above-discussed graphics may includethe following the ability to highlight and track elements in theplot, the capacity to control and measure the speed of the componentsand the ability to control the parameters that compare differentscenarios.
Theresearchers who are interested in this field are welcomed tocontribute their ideas that will enable the data visualizationprocess to be expanded. Also, the parties which are interested areadvised to give their views on the areas that are supposed to bescraped off or incorporated in this research tool.
Carroll, N. D. (2014). Visualization and analytics tools for infectious disease epidemiology. Journal of biomedical informatics, 322-355.
Luz, P. Z. (2015). CD4 response up to 5 years following combination antiretroviraltherapy in HIV-infected patients in Latin America. The Caribbean Open Forum Infectious Diseases, 33-45.
Meridith, B. F. (2016, March 10). Leveraging Data Exchange Standards to Share and Reuse ResearchTools. Interactive Data Visualization for HIV Cohorts, pp. 1-11.
Robinson, A. M. (2011). Designing a web-based learning portal for geographicvisual- ization and analysis in public health. Health informatics journal, 191-208.