VISUALISAS

I ANALISIS
DATA
SPASIAL
Noor Alis Setiyadi

This Photo by Unknown author is licensed under CC BY-NC.

Tujuan pembelajaran

• Mahasiswa mampu mengenal visualisasi data spasial dan perannya dalam kesehatan
masyarakat
• Mahasiswa menemukan visualisasi data informasi penting, seperti kurva epidemi (atau
epi), menyampaikan tentang penyebaran penyakit.
• Mahasiswa akan menggunakan data surveilans dari fiksi, novel emerging respiratory
disease (NERD) untuk membuat kurva epi, mengidentifikasi pola yang mengungkapkan
kemungkinan mode penularan, dan membuat rekomendasi strategi pencegahan untuk
mengurangi penyebaran penyakit.
 How are public health data
visualized?

https://www.cdc.gov/scienceambassador/nerdacademy/data-interpretation.html
 • Perhatikan paparan video ttg
Skenario bagaimana data kesmas
pembelajaran divisualisasikan di
pertemuan 2 https://youtu.be/I-tH_juih4w
• Yang bisa dipetik dari video presentasi tersebut adalah visualisasi data
sangat penting dalam mengkomunikasikan informasi lebih mudah
dipahami.
• Disamping itu, dapat melihat trend, pola, dan kemungkinan
kemungkinan dari fakta data Kesehatan yang dijelaskan.
 Hanya menampilkan angka

https://amcrisan.github.io/assets/files/slides/2015-05-BCCDC-Rounds-PH-Data-Vis_v2.pdf
Evidence from Risk Communication Literature
(difficult to understand) (easier to understand)
Probability < Frequency < Visualization
60% 6 in 10

Numeracy : the ability to reason with numbers

Individuals with low numeracy have a difficulty interpreting numbers and

probabilities

Visualizations can help people with low numeracy make sense of data,

But, there is some evidence that low numeracy affects reasoning with graphs
as well.
Whiting et. al (2015) “How well do health professionals interpret diagnostic information? A systematic review”

14
Tampilan angka dan gambar
• Example
Report:
OncotypeDx
DCIS report
Tampilan angka dan gambar
• Example
Report: Myriad
Prolaris
Prostate Cancer
Test Report
Tampilan angka dan gambar
• Example Report:
Decipher Prostate
Cancer Test Report

•Primary population:
•Men, who are
susceptible to red-
green colour
blindness
• TB incidence rates overlain on geography

•Communicate rates of TB by Health Service

Delivery Area (HSDA) region
• John Snow’s Visualization
of the 1854 Cholera
Outbreak
Example : Visualizing Public Health Data

34
A Data visualization in 3 Questions:

Why? (Motivation)
Why do you need to visualize data?

What? (Data)
What kind of data is being visualized?

How? (Visual and Interaction Design)

How is data being visualized?

35
A Data visualization in 3 Questions:

Design Evaluation
Does the visualization solve a
Why? relevant problem?

Are you using the right data, or

What? deriving the right data?

How? Are the visual and interactive

design choices appropriate?
36
Steps to Design and Evaluate a Data Visualization

Why

What
DESIGN
How

How

EVALUATIO
N

Munzner (2014) “Visualization Analysis and Design” 37

Steps to Design and Evaluate a Data Visualization
Methodology
Qualitative Why
Methods,
Domain Knowledge What
Design &
How
Cognitive Science
How
Computer Science

Qualitative &
Quantitative
Methods

38
Part 2:
How I plan to answer the question

39
How Data Visualization is like Statistical Modelling
Model selection is a design problem

statistical model

Input data Parameters

(to fit the
model)
40
How Data Visualization is like Statistical Modelling
“Parameters” of Visual and Interaction Design!
Colour = Continent

Transparency = Similarity

Size = Population

Five dimensions are plotted in 2D

(4 continuous dimensions & 1 categorical 41
Basic Building Blocks of Data Visualization
“Parameters”

Munzner (2014) “Visualization Analysis and Design”

42
How Data Visualization is like Statistical Modelling
“Parameters” of Visual and Interaction Design!
Colour = Continent

Transparency = Density

Reveal detail on hover

43
How Data Visualization is like Statistical Modelling
“Parameters” of Visual and Interaction Design!

The same parameters can be

combined in different ways to
yield different visualizations

44
How Data Visualization is like Statistical Modelling
A finale note on parameters

For brevity, I haven’t exhaustively described all the different

components, which I’ve called parameters, that can be a
part of data visualization

For more in depth details consider:

Visualization Design and Analysis (2014) by Tamara Munzner

45
How Data Visualization is like Statistical Modelling
OPTIMIZATION!

Finding the best model

Searching the parameter
space for a model that
error

yields that lowest error

46
How Data Visualization is like Statistical Modelling
OPTIMIZATION!
The “Design Space” metaphor

Sedlmair 2012
https://www.cs.ubc.ca/nest/imager/tr/2012/dsm/dsm-talk.pdf
47
How Data Visualization is like Statistical Modelling
OPTIMIZATION!
The “Design Space” metaphor

Sedlmair 2012
https://www.cs.ubc.ca/nest/imager/tr/2012/dsm/dsm-talk.pdf
48
Progressively Identify the Right Visualization
The “Design Space” metaphor

Use “why, what, and how” framework to guide the selection

of the optimal design choice
Sedlmair 2012
https://www.cs.ubc.ca/nest/imager/tr/2012/dsm/dsm-talk.pdf
49
The Importance of Thinking Broadly

Use “why, what, and how” framework to guide the selection

of the optimal design choice
Munzner (2014) 'Visualization Analysis and Design" SO
Designs for Visualizing Health Data (http://www.vizhealth.org/)

S1
How Data Visualization is like Statistical Modelling
A final note

Data visualization and statistical modelling are not

identical, even though at a high-level they share similar
research processes

I’ve presented one aspect of visualization research, but

there are others I haven’t touched upon

I’ve emphasize problem driven work – finding the right visualization

for a specific motivation or task – but there also exists technique
and systems type research

S2
How to Implement Data Visualizations

Matthew Brehmer’s totally subjective ranking of vis design tools

S3
BUT…..
How do we
design good
visualizations
for public
health?
S7
Motivations Underlying my Doctoral Work

Decision Support Design Space

For communicable disease Characterizing and evaluation
prevention and control the design space of public
health microbial genomics
Motivations Underlying my Doctoral Work
Methodology
Desig ning and evaluating data
visual izations through a public h ealth lens

Decision Support Design Space

For communicable disease Characterizing and
prevention and control evaluation the design space
of public health microbial
genomics
DECISION SUPPORT
Visualizing Tuberculosis data at the
British Columbia Centre for Disease Control

S7
WHY
S8
Combining Data will Prepare us for the
Pandemics of the Future

La
b

Social

Cli nical

S9
But, that’s a lot of data….

6O
Can Visualizing TB data help Decision Support?
We wanted to create an interactive and visual tool that allowed our
public health stakeholders to analyze the different data types
We want to understand how this tool can be used by different
public health stakeholders

Medical Health Officers TB Clinicians TB Nurses Researchers Epis / Biostats

61
WHAT
62
Treatment

Outcomes

Genomic

Patient Data
Geography /
Contact Network
Location

63
Treatment
Genomic
TB whole genome Outcomes
Genotyping

Patient Data
Geography /
Contact Network
Location

67
 Treatment

Outcomes

Genomic

Patient Data
Geography /
Contact Network
Location
time
6S
HOW
66
An It rative Approach to Development

An iterative approach to development allows us to get

feedback before committing to ineffective design choices
67
 The Big Picture
But this takes a lot of time & effort Effort
most

Latent TB
Patient Demographic & Treatment Data

Active TB
Genomic (Whole TB Genome Sequencing Data)

Active TB
Genomic (MIRU- VNTR) & Contact Data

Active TB
Patient Demographic &
Treatment Data

Time
least
most
68
 Introducing EpiCOGs

DEMO

EpiCogs is a data viewer and currently a sandbox

environment for developing data visualizations 69
Factors Influencing the Current Design
Task: Filter patients and identify where they are
Filter patients from the side panel, and interactively update
the line list & map based upon those interactions
Task: Follow-up on selected patients
Select patients view – a subset of the data. For out reach
nurses, and request to include driving directions.
Task: Incorporate existing statistical methods
Analysis modes, allows epidemiologists and
biostatisticians to integrate their R methods into EPI COGS
Task: Provide overview of key metrics
Predefined analysis modules that in the future will be
migrated to “reports” section.
Factors Influencing the Current Design
Needs of individuals
Gathered through meetings, dialogue with individuals, and
various iterations of EpiCOGs
Technology Changes
Support for data visualization tools in R improved greatly
allowing for the creation of better data visualizations
Data Driven Interface and Analysis
Created a data driven interface that is responsive to the
user’s data.
Policies and Procedures
Existing policies and procedures at the BCCDC inform the
utility of such a tool and how it can integrate into existing
workflows
Initial Work & Next Directions
Much initial work was to understand the tool’s feasibility
Could it meet the needs of stakeholders?
How could it integrate (security & workflow)?
How could it be supported long term? (Choice of R)
Could we build a useful tool in R?

Next phases will explore genotypes, genomics, and

contact networks
Right now, users can filter based on assigned genotype
clusters (which will show patients on map), but we’re working
towards better visual and interactive design for these data