EventFlow: Visual Analysis of Temporal Event Sequences
and Advanced Strategies for Healthcare Discovery
Jump to: Features | Participants | Presentations | Videos | Software | User Support | Publications | Press | Related Projects
- See also our latest project: CoCo for Cohort Comparison
- EventFlow is now available for Commercial Licensing (see below)
- Our book chapter on Information Visualization - including a case study of the use of EventFlow for the analyis of asthma medication prescriptions - appears as Chapter 12 in
Big Data and Health Analytics, Katherine Marconi and Harold Lehman (Eds), CRC Press - Taylor and Francis (2014).
- Join us in Paris for the IEEE VIS Workshop on Visualization fof EHR data, Paris, France, November 9, 2014
- Strong participation of our EventFlow users at the 2014 Workshop on Visualization of Temporal Patterns in EHR data:, in association with the Annual HCIL Symposium, May 29, 2014, College Park.
- Oracle produces a video highlighting our work.: January 2013
EventFlow: Advanced Strategies for Healthcare Discovery
While Eventflow has been used in a variety of other application domains (cybersecurity, sports analytics, incident management etc.) the analysis of healthcare data has been our driving application domain.
With Eventflow health providers, insurance companies, and hospitals now have a powerful tool for discovering patterns in Electronic Health Records and claims reports. EventFlow displays point and interval events in
(1) a visually comprehensible format for each patient, as well as in
(2) a novel aggregation view that reveals common and rare patterns across the entire database.
The innovative graphical user interfaces offers powerful search capabilities that enable users to specify temporal patterns. The ranked result sets using the visually comprehensible format for each patient. Search & Replace features, plus rich manipulations enable users to transform their datasets so as to confirm required treatment patterns, detect missing events, and reveal unexpected sequences.
Medical researchers have already applied EventFlow to analyze treatment patterns and outcomes, while network security analysts have studied cyberattack patters and sports analysts have found novel approaches to studying games, overall team performance, and seasonal patterns. Applications for web log, sensor data, business processes, and financial transactions constitute secondary markets.
The HCIL's ongoing work with temporal event records has produced powerful tools for analyzing and exploring patterns of point-based events (Lifelines2, LifeFlow). However, users found that point-based events limited their capacity to solve problems that had inherently interval attributes, for example, the 3-month interval during which patients took a medication. To address this issue, EventFlow extends its predecessors to support both point-based and interval-based events. Interval-based events represent a fundamental increase in complexity at every level of the application, from the input and data structure to the eventual questions that a user might ask of the data. Our goal was to accomplish this integration in a way that appeared to users as a simple and intuitive extension of the original LifeFlow tool. With EventFlow, we present novel solutions for displaying interval events, simplifying their visual impact, and incorporating them into meaningful queries.
- Megan Monroe, PhD Candidate, Computer Science
- Sana Malik, PhD Candidate, Computer Science
- Fan Du, PhD Candidate, Computer Science
- Christopher Imbriano , PhD Candidate, Computer Science
- Catherine Plaisant, Research Scientist, UMIACS
- Ben Shneiderman, Professor, Computer Science
- Jeff Millstein, Oracle Research Past Participants
- Rongjian Lan, PhD Candidate, Computer Science
- Krist Wongsuphasawat PhD, Computer Science
- Sigfried Gold, Social & Scientific Systems Collaborators
- We appreciate the collaboration of clinical researchers and epidemiologists at the US Army Pharmacovigilance Center, University of Maryland School of Pharmacy, National Children Hospital, University of Florida, Washington Hospital Center, and many others.
We appreciate the partial support of Oracle Corporation and the Center for Health-related Informatics and Bioimaging (CHIB) at the University of Maryland. EventFlow builds on early work which was funded in part by NIH - National Cancer Institute grant RC1-CA147489 "Interactive Exploration of Temporal Patterns in Electronic Health Records" (for LifeLines2 and LifeFlow), and later by the Maryland Industrial Partnerships (MIPS) program and Pulse8.
If you are in a rush, this video provides a good overview of how the EventFlow aggregation is constructed, and how the search and replace can be used to manipulate the data to answer questions.
Or: take a look at the analysis of BASKETBALL data:
The following slides provide an introduction to the motivation behind EventFlow, and a summary of its features. A full description can be found in the tech report.
We can provide you with a review version of Eventflow
- For non-commercial use: please contact firstname.lastname@example.org with a description of your project and organization)
- For commercial use: EventFlow is available for licensing. To request a review copy of EventFlow and for more information about licensing please contact:
Office of Technology Commercialization (OTC)
2130 Mitchell Building, University of Maryland, College Park, MD 20742
Phone: 301-405-3947 | Fax: 301-314-9502
- Not sure: contact email@example.com
Users can watch the videos above, or refer to the EventFlow Manual, which covers all the major features of EventFLow.
To learn more about the EventFlow data format, go to the EventFlow data format section in the user manual.
To generate datasets for testing, use the Dataset Generator.
Suggestions for dealing with large datasets: Sharpening Analytic Focus to Cope with Big Data Volume and Variety: Ten strategies for data focusing with temporal event sequences
SHORT PAPER TO GET STARTED: A case study of our collaboration effort with the US Army Pharmacovigilance Center
Megan Monroe, Tamra Meyer, Catherine Plaisant, Rongjian Lan, Krist Wongsuphasawat, Trinka Coster, Sigfried Gold, Jeff Millstein, Ben Shneiderman
A Pilot Study of Asthma Medications in the Military Health System
A much more detailed version of the case study also appeared in Chapter 12 of Big Data and Health Analytics, Katherine Marconi and Harold Lehman (Eds), CRC Press - Taylor and Francis (2014)
[BEST REFERENCE] VAST2013 paper describing EventFlow and the simplification of temporal event sequences
Megan Monroe, Rongjian Lan, Catherine Plaisant, Ben Shneiderman
Temporal Event Sequence Simplification
Proceedings of IEEE VAST 2013 Conference.
[GOOD REFERENCE] CHI 2013 Paper on EventFlow's interactive QUERY approach
Megan Monroe, Rongjian Lan, Juan Morales del Olmo, Catherine Plaisant, Ben Shneiderman, and Jeff Millstein
The challenges of specifying intervals and absences in temporal queries:A graphical language approach.
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems 2013
Another case study short paper
Carter, E., Burd, R., Monroe, M., Plaisant, C., Shneiderman, B.
Using EventFlow to Analyze Task Performance During Trauma Resuscitation
Proc. of the Workshop on Interactive Systems in Healthcare, WISH2013 (2013)
Paper on temporal query processing algorithms
Megan Monroe, Amol Deshpande
An Integer Programming Approach to Temporal Pattern Matching Queries
Unpublished Tech Report: early work on temporal event sequence search and replace
Rongjian Lan, Hanseung Lee, Megan Monroe, Allan Fong, Catherine Plaisant, Ben Shneiderman
Temporal Search and Replace: An Interactive Tool for the Analysis of Temporal Event Sequences
Unpublished Tech Report: Early report on extending the Lifeflow overview visualization to handle interval data
Megan Monroe, Krist Wongsuphasawat, Catherine Plaisant, Ben Shneiderman, Jeff Millstein and Sigfried Gold
Exploring Point and Interval Event Patterns: Display Methods and Interactive Visual Query
Other publications written by our case study partners
E. Onukwugha, Y. Kwok, C. Yong, C. Mullins, B. Seal, A. Hussain,
Variation in the length of radiation therapy among men diagnosed with Incident Metastatic Prostate Cancer"
Poster presented at the 2013 ASTRO (American Society for Radiation Ocology)meeting.
See also our 2014 HCIL Workshop on Visualization of Temporal Patterns in EHR data:, that included several presentations from EventFlow users.
General survey paper
Rind, A., Wang, T., Aigner, W., Miksch, S., Wongsuphasawat, K., Plaisant, C., Shneiderman, B., Interactive Information Visualization for Exploring and Querying Electronic Health Records: A Systematic Review, in Foundations and Trends in Human-Computer Interaction, Vol. 5, No. 3 (2013) 207-298.
Hunter Whitney, It's About Time, UX Magazine, September 2014.
Products and papers stimulated by this work
Our original research on LifeFlow and EventFlow stimulated new work by other labs, such as:
OutFlow and CareFlow at IBM. Visualizing Uncertain Critical Paths in Schedule Management, by Robert Gove, Brandon Herzog (see VAST 2013 Industry posters)
Related projects and events
Strong HCIL participation at the Visual Analytics in Healthcare (VAHC 2013) workshop at AMIA, with Ben Shneiderman's keynote and 2 papers and demos. November 2013
Links of all HCIL Projects related to Temporal Visualization: LifeLines, LifeLines2, PatternFinder, LifeFlow, etc.
HCIL EHR Informatics Workshop 2013: Thursday, May 23, 2013
EventFlow: User Group Meeting: Tuesday, November 20, 2012
HCIL EHR Informatics Workshop 2012: Wednesday, May 23, 2012