Definition of Visual Analytics
Visual analytics is the science of analytical reasoning facilitated by interactive visual interfaces (Thomas, 2006). Visual analytics tools can enable a user to construct simple, easy-to-use dashboards that display histograms and graphs in order to identify data trends. This makes the data accessible in a way that facilitates understanding—one of the most important goals of the application of analytics to massive amounts of data (Barton 2012). Visual analytics is used for data exploration and hypothesis generation within a specific group of data, and is often the step prior to the application of advanced data analytics.
Visual Analytics in Health Care
Visual analytics has been utilized to enhance the evaluation of large, complex data sets not only within health care, but also in a variety of health care-related fields, such as genomics, immunology, and epidemiology (Kumasaka, 2010; Naumova, 2010; Chui, 2011). Furthermore, the massive data repositories that are generated by electronic health records (EHRs) can be navigated and represented visually in near-real time by the use of visual analytics tools
The use of visual analytics in health care settings usually can be categorized of one of three types: clinical operations (e.g. blood bank, hand hygiene analysis), scientific (e.g. research data analysis), and business (e.g. supply chain, meaningful use). There are multiple visual analytics software products that are available for purchase; popular examples include Tableau and Qlikview .
Many examples of the use of visual analytics in health care data analysis are available in the literature, such as for tracking symptom evolution during disease progression or pharmacokinetic-pharmacodynamic analysis (Goldsmith, 2010; Perer, 2012). Visual analytics can also facilitate investigative analysis in research by showing connections between entities, focusing on essential information, and reviewing hypotheses (Kang, 2010).
Wang, et al (2011) described one example of the successful application of visual analytics for data exploration in health care. The research group used a visual analytics application, Lifelines2, to providers in a variety of specialties (e.g. neurologist, osteopathic physician, emergency room director) to visualize EHR data with the goal of improving patient care. The authors described the providers’ use of their tool for tasks that were difficult to answer with the providers’ EHR software: studying hospital room transfer patterns, performing follow-up studies and replicating studies, among others. Of particular interest was the ability of their data visualization tool to visualize data from multiple EHRs, as many of the visual analytics tools described in the literature present a dashboard displaying a particular data set.
Visual analytics tools can also be used for data visualization to provide clinical decision support at the point of care. Mane (2012) described a “VisualDecisionLinc” (VDL) tool that displayed near real-time comparative population evidence generated from their institution’s EHR data in order to provide clinical decision support in psychiatric patients. The tool enabled the generation of EHR-data-based dynamic charts that helped clinicians weigh the risks of therapeutic options and outcomes.
The future of visual analytics in health care will consist of an ever-increasing application of dashboards and tools to visualize and analyze data in order to improve patient care, increase efficiency, facilitate resource utilization and allocation, and enhance decision-making.
Barton D, Court D. Making advanced analytics work for you. Harv Bus Rev. 2012 Oct; 90(10):78-83, 128.
Chui KK, Wenger JB, Cohen SA, Naumova EN. Visual analytics for epidemiologists: understanding the interactions between age, time, and disease with multi-panel graphs. PLoS One. 2011 Feb 15;6(2).
Goldsmith MR, Transue TR, Chang DT, Tornero-Velez R, Breen MS, Dary CC. PAVA: physiological and anatomical visual analytics for mapping of tissue-specific concentration and time-course data. J Pharmacokinet Pharmacodyn. 2010 Jun;37(3):277-87.
Kang YA, Görg C, Stasko J. How can visual analytics assist investigative analysis? Design implications from an evaluation. IEEE Trans Vis Comput Graph. 2011;17(5):570-83.
Kumasaka N, Nakamura Y, Kamatani N. The textile plot: a new linkage disequilibrium display of multiple-single nucleotide polymorphism genotype data. PLoS One. 2010;5(4)e10207.
Mane KK, Bizon C, Schmitt C, Owen P, Burchett B, Pietrobon R, Gersing K. VisualDecisionLinc: a visual analytics approach for comparative effectiveness-based clinical decision support in psychiatry. J Biomed Inform. 2012 Feb;45(1):101-6.
Naumova EN. Visual analytics for immunologists: Data compression and fractal distributions. Self Nonself. 2010 Jul;1(3):241-249.
Perer A, Sun J. MatrixFlow: temporal network visual analytics to track symptom evolution during disease progression. AMIA Annu Symp Proc. 2012;2012:716-25.
Thomas JJ, Cook KA. A visual analytics agenda. IEEE Comput Graph Appl. 2006 Jan-Feb;26(1):10-3.
Wang TD, Wongsuphasawat K, Plaisant C, Shneiderman B. Extracting insights from electronic health records: case studies, a visual analytics process model, and design recommendations. J Med Syst. 2011 Oct;35(5):1135-52.
Submitted by Allan Simpao