A Visual Analytics Approach to Comparing Cohorts of Event Sequences - PowerPoint PPT Presentation

A Visual Analytics Approach to Comparing Cohorts of Event Sequences 
 Sana Malik, Fan Du, Megan Monroe, Eberechukwu Onukwugha, Catherine Plaisant, & Ben Shneiderman May 29, 2014 — HCIL Symposium

Time-stamped event data is widespread

Time-stamped event data is widespread → E-commerce

Time-stamped event data is widespread → E-commerce → Online education

Time-stamped event data is widespread → E-commerce → Online education → Patient medical histories

Often, we compare groups in these datasets

Often, we compare groups in these datasets → E-commerce

Often, we compare groups in these datasets → E-commerce

Existing tools fall into two categories.

1. Visual

Aspirin -> Emergency -> Normal Floor Bed -> ICU -> Exit 38.37, 0.0, 4.11e-123 Emergency -> ICU -> Exit 24.61, 0.0, 2.11e-73 Emergency -> Normal Floor Bed -> Exit -> ICU 2. Statistical 5.26, 0.0, 4.12e-15 Emergency -> Normal Floor Bed -> ICU -> Exit 5.26, 0.0, 4.12e-15 Aspirin -> Emergency -> ICU -> Normal Floor Bed -> Exit 7.89, 3.22, 1.20e-06 Aspirin -> Emergency -> ICU -> Intermediate Care -> Exit 4.15, 0.0, 4.22e-12 Emergency -> Normal Floor Bed -> ICU -> Intermediate Care -> Exit 2.97, 0.0, 7.02e-09 Aspirin -> Emergency -> ICU -> Exit 2.80, 0.0, 2.02e-08 Emergency -> Normal Floor Bed -> ICU -> Normal Floor Bed -> Exit 1.44, 0.0, 9.83e-05 Emergency -> Normal Floor Bed -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> Exit 1.01, 0.0, 0.00 Aspirin -> Emergency -> Normal Floor Bed -> ICU -> Intermediate Care -> Exit 0.76, 0.0, 0.00 Aspirin -> Emergency -> Normal Floor Bed -> Exit 0.0, 1.61, 3.37e-05 Aspirin -> Emergency -> ICU -> Normal Floor Bed -> ICU -> Exit 0.0, 2.37, 2.84e-07 Aspirin -> Emergency -> Normal Floor Bed -> ICU -> Normal Floor Bed -> Exit 3.82, 6.45, 0.00 Emergency -> ICU -> Normal Floor Bed -> ICU -> Exit 0.0, 4.07, 7.17e-12 Emergency -> ICU -> Normal Floor Bed -> ICU -> Normal Floor Bed -> Exit 0.0, 4.24, 2.48e-12

visualization + statistics = CoCo (Cohort Comparison)

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results Emergency Normal Floor Bed

Visualizing Statistical Results (59.8%) (10.1%) Emergency Normal Floor Bed Δ = 49.7% (p = 0.001)

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results

Visualizing Statistical Results

1. Summary Statistics

2. Prevalence Metrics → Events

2. Prevalence Metrics → Events → Event sequences

2. Prevalence Metrics → Events → Event sequences → Co-occurring (non-sequential) events

2. Prevalence Metrics → Events → Event sequences → Co-occurring (non-sequential) events → Outcomes of records

3. Time Metrics → Absolute times

3. Time Metrics → Absolute times → Duration of events, gaps, and overlaps

4. Attribute Metrics → Event attributes Emergency | Doctor = Smith

4. Attribute Metrics → Event attributes → Record attributes

Pilot User Study → 10 participants → Used CoCo & side-by-side EventFlow for analysis → Counted the number of insights users made

Pilot User Study

Pilot User Study “ I like the fact that you give me two different tools. I can look at the data in different ways ” ! → 9/10 participants started with EventFlow for an “overview” → Used CoCo to confirm findings

Future Work → Support a variety of statistical tests → Add visualizations of distributions → Support exploration and search of sequences

A Visual Analytics Approach to Comparing Cohorts of Event Sequences 
 Sana Malik, Fan Du, Megan Monroe, Eberechukwu Onukwugha, Catherine Plaisant & Ben Shneiderman For more information & to become a CoCo user: web www.cs.umd.edu/hcil/coco email maliks@cs.umd.edu ! ! ! Supported by Oracle and the University of Maryland Center for Health-related Informatics and Bioimaging (CHIB), a unit of UMIACS