Event Analytics and the Visualization of Temporal Event Sequences - - PowerPoint PPT Presentation

▶
event analytics and the visualization of temporal event
SMART_READER_LITE
LIVE PREVIEW

Event Analytics and the Visualization of Temporal Event Sequences - - PowerPoint PPT Presentation

Mar 22, 2023 278 likes •813 views

Event Analytics and the Visualization of Temporal Event Sequences Catherine Plaisant plaisant@c.umd.edu Journes Vis~ June 8, 2017 ENSAM & Jenny Preece Ben Shneiderman Research collaborator for 30 years Information visualization

slide-1
SLIDE 1

Event Analytics and the Visualization of Temporal Event Sequences

Catherine Plaisant

plaisant@c.umd.edu Journées Vis~ June 8, 2017

slide-2
SLIDE 2

ENSAM

slide-3
SLIDE 3
slide-4
SLIDE 4
slide-5
SLIDE 5

& Jenny Preece

Ben Shneiderman Research collaborator for 30 years

slide-6
SLIDE 6

Information visualization

Hierarchical Clustering Explorer HCE Home Finder and Filmfinder prototypes lead to SpotFire Treemap SpaceTree Lifelines

slide-7
SLIDE 7

Event Analytics

slide-8
SLIDE 8

Patient ID: 45851737

12/02/2008 14:26 Arrival 12/02/2008 14:36 Emergency 12/02/2008 22:44 ICU 12/05/2008 05:07 Floor 12/14/2008 06:19 Exit

Time Emergency ICU Floor Exit

Numerical Categorical

04/26/2010 10:00 31.03 04/26/2010 10:15 31.01 04/26/2010 10:30 31.02 04/26/2010 10:45 31.08 04/26/2010 11:00 31.16

Patient ID: 12345

Arrival e.g. High/Normal/Low + intervals

slide-9
SLIDE 9

Prototype evolution

www.cs.umd.edu/hcil/toolname

Tool l Event nt Typ ypes Re Records Di Displa lay y LifeLines Points, Intervals One Individual LifeLines2 Points Many Individual, Summary Similan Points Many Individual LifeFlow Points Many Individual, Aggregate EventFlow Points, Intervals Many Individual, Aggregate

slide-10
SLIDE 10

Electronic Health Records: symptoms, treatment, lab test Student records: course, paper, proposal, defense, etc. Web logs, usability logs, security etc. Traffic incident logs: confirmed, unit arrived, lane closed etc.

Many application domains

slide-11
SLIDE 11

A SINGLE RECORD

What is the situation? What has been done? What should we do now?

slide-12
SLIDE 12

MULTIPLE RECORDS

RECORD RECORD R E C O R D RECORD RECORD

Are we following guidelines? Do you have patients for my clinical trial? Retrospective analysis: How are opioids prescribed? Patterns of readmissions? Drug adverse reactions? How can we improve care?

slide-13
SLIDE 13

A SINGLE RECORD

slide-14
SLIDE 14

LifeLines – Single Patient

http://www.cs.umd.edu/hcil/lifelines

slide-15
SLIDE 15

1753

Jacques Barbeu Du Bourg - 1753

http://gallica.bnf.fr/ark:/12148/bpt6k1314025

slide-16
SLIDE 16

Controlled experiment

Alonso, D., Rose, A., Plaisant, C., Norman, K. Viewing Personal History Records: A Comparison of Tabular Format and Graphical Presentation Using LifeLines Behavior and Information Technology 17, 5, 1998, 249-262.

  • 36 participants used either LifeLines or Tabular display (static display only)
  • Series of tests:

first impression, 31 question quiz, subjective satisfaction questionnaire, recall test, spatial ability

slide-17
SLIDE 17
  • LifeLines twice as fast for many tasks

(e.g. time interval comparison, or task across categories)

  • Better recall

4.3 vs 2.8 correct out of 6 questions

  • More accurate “1st impression”

Strong benefits

slide-18
SLIDE 18

MULTIPLE RECORDS

RECORD RECORD R E C O R D RECORD RECORD

slide-19
SLIDE 19

Constructing the EventFlow Overview

A B C D E

slide-20
SLIDE 20
slide-21
SLIDE 21
slide-22
SLIDE 22
slide-23
SLIDE 23
slide-24
SLIDE 24
slide-25
SLIDE 25
slide-26
SLIDE 26
slide-27
SLIDE 27
slide-28
SLIDE 28
slide-29
SLIDE 29

Number of Records Time

slide-30
SLIDE 30

EventFlow ~ e.g. Transfers within Hospital

slide-31
SLIDE 31
slide-32
SLIDE 32

Children’s Hospital in DC: Trauma Bay

slide-33
SLIDE 33

33 33

Airway Breathing Circulation Disability Exposure

Primary Survey (ABCDE)

Identify and manage life-threatening conditions in a sequential manner

slide-34
SLIDE 34

34 34

Secondary Survey

Begins once Primary Survey is complete and patient is stable Head to Toe Examinati

  • n
slide-35
SLIDE 35

skip At first… Confetti. Need strategy to get answers

slide-36
SLIDE 36

Focusing on the first two events

84% of patients are checked in the correct order. The most common deviation is that the breathing is checked before the airways (14% of patients) Reversed group takes longer on average than the correct sequence

slide-37
SLIDE 37

Distributions

slide-38
SLIDE 38

Adding third event type (central pulse)

slide-39
SLIDE 39

Adding distal pulse Combine the 2 pulse types

slide-40
SLIDE 40

Graphical search & replace to remove duplicates

slide-41
SLIDE 41

81% of the patients are treated in the correct order. The largest deviation is still the airway and breathing being out of order, but there are also instances where the circulation is checked too early.

slide-42
SLIDE 42

78% of patients treated in the correct order…

Add disability check

slide-43
SLIDE 43

The average time between steps continues to be longer for the reversed airway and breathing check

  • groups. The time between circulation and disability checks is about 1 minute for the correct sequence and

2 minutes and 15 seconds for the reversed order group.

slide-44
SLIDE 44

Searching with a time constraint.

See where check is taking longer than one minute

> 1 min

slide-45
SLIDE 45

Add secondary survey

Correct procedure drops to only 48% of patients,

slide-46
SLIDE 46

Other example

in collaboration with Army PharmacoVigilance Center Analysis of prescription patterns of asthma medication

slide-47
SLIDE 47

20 case studies

slide-48
SLIDE 48

20 case studies 15 strategies

Du et al. (2017) Coping with Volume and Variety in Temporal Event Sequences: Strategies for Sharpening Analytic Focus IEEE Transactions Visualization and Computer Graphics

slide-49
SLIDE 49

Event Analytics Eventflow to: review the data from individual records search for temporal patterns of interest summarize all the event sequences perform data transformations select cohorts of interest for further studies

see also lots of other projects in event analytics e.g. workshop at IEEE Vis 2016

slide-50
SLIDE 50

Merci

Thank you to U. of Maryland colleagues Ben Shneiderman Fan Du, Sana Malik Megan Monroe, Krist Wongsuphasawat, David Wang + all case study partners

plaisant@cs.umd.edu hcil.umd.edu/eventflow

Contact me for software access FYI: full day course June 30th

www.aviz.fr/DayCourse2017/EventAnalytics

slide-51
SLIDE 51

Common question: Comparison of 2 groups of records. Two Eventflow side by side (hard to find differences)

slide-52
SLIDE 52

Cohort Comparison: Coco

www.cs.umd.edu/hcil/coco

slide-53
SLIDE 53

Cohort Comparison: Coco

www.cs.umd.edu/hcil/coco