MetricScalpel: Analyzing Diagnostic Outcomes with Exploratory Data Visualization

Healthcare data is the emerging force to push ahead smarter clinical solutions. However, unforeseen challenges like unmanaged massive data and costly access to it make it difficult for domain experts to easily derive actionable insights. In collaboration with a local diagnostic service provider, we designed MetricScalpel, a web-based visualization tool to help people quickly look into the real-life diagnostic data in the local community. The tool enables swift exploration into the multivariate diagnostic data sets from different sources and facilitates data selection/subsetting for deeper analysis. It can be used to reveal overlooked health conditions on almost any level without the requirement of heavy technical knowledge. Such design makes it easier to be accepted by a wider user group in the healthcare related organizations. It was proven to well serve the domain experts in validating pre-exist hypotheses in cohort analysis as well as revealing undiscovered patterns of health conditions in the local community. External evaluation shows operation cost was remarkably confined as domain experts were assisted with direct and intuitive access to the relevant data in need.

[1]  Jeffrey Heer,et al.  Enterprise Data Analysis and Visualization: An Interview Study , 2012, IEEE Transactions on Visualization and Computer Graphics.

[2]  E. AbuKhousa,et al.  Predictive data mining to support clinical decisions: An overview of heart disease prediction systems , 2012, 2012 International Conference on Innovations in Information Technology (IIT).

[3]  Rainu Kaushal,et al.  Trends in public perceptions of electronic health records during early years of meaningful use. , 2015, The American journal of managed care.

[4]  Nele Verbiest,et al.  Risk-O-Meter: an intelligent clinical risk calculator , 2013, KDD.

[5]  Mario Delgado-Noguera,et al.  Factors associated with length of hospital stay in minor and moderate burns at Popayan, Colombia. Analysis of a cohort study. , 2016, Burns : journal of the International Society for Burn Injuries.

[6]  D. Sharma,et al.  A Security Architecture for e-Health Services , 2008, 2008 10th International Conference on Advanced Communication Technology.

[7]  Ben Shneiderman,et al.  Cohort Comparison of Event Sequences with Balanced Integration of Visual Analytics and Statistics , 2015, IUI.

[8]  Susan Clamp,et al.  Health informatics education for clinicians and managers - What's holding up progress? , 2004, Int. J. Medical Informatics.

[9]  Zhang Ning,et al.  A solution for an application of information visualization in telemedicine , 2012, 2012 7th International Conference on Computer Science & Education (ICCSE).

[10]  Lisette Schoonhoven,et al.  Development of an e-supported illness management and recovery programme for consumers with severe mental illness using intervention mapping, and design of an early cluster randomized controlled trial , 2015, BMC Health Services Research.

[11]  John Hsu,et al.  Early Experiences with e-Health Services (1999–2002): Promise, Reality, and Implications , 2006, Medical care.

[12]  W. Liu,et al.  e-Health security context management service , 2015, 2015 International Conference on Computing, Networking and Communications (ICNC).

[13]  Zhiyuan Zhang,et al.  Iterative cohort analysis and exploration , 2015, Inf. Vis..