Embedded Files
CS 424 Student's Choice
By: Jack Martin
Fig 1. Overview and Initial landing page of FluView
FluView's Purpose
FluView's Purpose
The purpose of this interactive visualization is to provide weekly influenza surveillance information in the United States and its respective jurisdictions. This application allows users to display and query information for current and previous influenza seasons. The scope of this data ranges from 1997 to 2022. These influenza data visualizations allow for broader patterns to be seen in order to prepare proper precautions to be taken against specific influenza strains such as vaccinations.
The purpose of this interactive visualization is to provide weekly influenza surveillance information in the United States and its respective jurisdictions. This application allows users to display and query information for current and previous influenza seasons. The scope of this data ranges from 1997 to 2022. These influenza data visualizations allow for broader patterns to be seen in order to prepare proper precautions to be taken against specific influenza strains such as vaccinations.
Fig 2. Menu to choose which source to visualize data from
What is the data?
What is the data?
There are two sets of data that can be selected to visualize, clinical labs and public health labs. The data that’s being collected from clinical laboratories include the weekly total number of specimens tested, the number of positive tests, and it also contains the percent positive categorized by the influenza type. The data that’s collected from public health laboratories include the weekly total of specimens tested, the number of positive tests, and this number by the type of influenza, both influenza A and B
There are two sets of data that can be selected to visualize, clinical labs and public health labs. The data that’s being collected from clinical laboratories include the weekly total number of specimens tested, the number of positive tests, and it also contains the percent positive categorized by the influenza type. The data that’s collected from public health laboratories include the weekly total of specimens tested, the number of positive tests, and this number by the type of influenza, both influenza A and B
How was this data collected?
How was this data collected?
This data was collected by both 110 World Health Organization (WHO) Collaborating laboratories and 270 National Respiratory and Enteric Virus Surveillance System (NREVSS). The U.S Outpatient Influenza-like Illness Surveillance Network (ILINet) which is composed of 3,200 outpatient healthcare providers all over the country and about 2,000 of these providers report data to the CDC on the number of patients they’ve seen and the amount who displayed influenza-like symptoms. They also report this based on the age group each patient resides in.
This data was collected by both 110 World Health Organization (WHO) Collaborating laboratories and 270 National Respiratory and Enteric Virus Surveillance System (NREVSS). The U.S Outpatient Influenza-like Illness Surveillance Network (ILINet) which is composed of 3,200 outpatient healthcare providers all over the country and about 2,000 of these providers report data to the CDC on the number of patients they’ve seen and the amount who displayed influenza-like symptoms. They also report this based on the age group each patient resides in.
Who are the users that this visualization was made for?
Who are the users that this visualization was made for?
The users that this visualization was made for medical professionals and also the general public. This is intended to improve the communication between healthcare professionals such as clinicians and scientists and the average citizen. With this communication, users can see relevant information about flu cases, their trends, and what particular strains are active.
The users that this visualization was made for medical professionals and also the general public. This is intended to improve the communication between healthcare professionals such as clinicians and scientists and the average citizen. With this communication, users can see relevant information about flu cases, their trends, and what particular strains are active.
What questions do users want to ask?
What questions do users want to ask?
Here are some potential questions that a user might want to ask:
Here are some potential questions that a user might want to ask:
- What is the most prevalent strain of influenza over the whole country during the 2015-2016 season?
a. Illinois?
a. Illinois?
2. Is the national ILI elevated or normal?
2. Is the national ILI elevated or normal?
3. Has the time period of the pandemic (2019-2022) increased or decreased the percentage of ILI related visits compared to a "normal" season like 2015-2016?
3. Has the time period of the pandemic (2019-2022) increased or decreased the percentage of ILI related visits compared to a "normal" season like 2015-2016?
4. Does the East North Central division have more cases of influenza than the New England division during 2012-2013?
4. Does the East North Central division have more cases of influenza than the New England division during 2012-2013?
a. Mid-Atlantic?
a. Mid-Atlantic?
5. When do the most cases of influenza happen every year?
5. When do the most cases of influenza happen every year?
Answers
Answers
What is the most prevalent strain of influenza over the whole country during the 2015-2016 season?
What is the most prevalent strain of influenza over the whole country during the 2015-2016 season?
Fig 3. Pie chart and stacked bar chart of the cumulative 2015-2016 season on a national level
Fig 4. Menu options to produce Fig 3.
The most prevalent strain of influenza over the United States for the 2015-2016 season is the H1N1 strain of the A variety of influenza. The steps that a user would have to get this information would be to select the 2015-2016 season, set viral surveillance to Public Health Labs in order to have more granularity between each strain, and then set the surveillance area to the National scale. Finally, they would have to set the pie chart to cumulative in order for it to contain all the cases in the season they're looking for.
The most prevalent strain of influenza over the United States for the 2015-2016 season is the H1N1 strain of the A variety of influenza. The steps that a user would have to get this information would be to select the 2015-2016 season, set viral surveillance to Public Health Labs in order to have more granularity between each strain, and then set the surveillance area to the National scale. Finally, they would have to set the pie chart to cumulative in order for it to contain all the cases in the season they're looking for.
For Illinois?
For Illinois?
Viral surveillance by Public Health Labs
Viral surveillance by Public Health Labs
H1N1 still seems to be the leading strain of influenza in Illinois
Viral surveillance by Clinical Labs
Viral surveillance by Clinical Labs
Clinical labs only report type A or B but remains consistent with the Public Health Labs
Is the national ILI elevated or normal?
Is the national ILI elevated or normal?
During the same time period as above, the national ILI is elevated at various points during the year. During the end of 2015 until the 15th week of 2016, the ILI is elevated compared to the national baseline of 2.1% of visits for ILI
During the same time period as above, the national ILI is elevated at various points during the year. During the end of 2015 until the 15th week of 2016, the ILI is elevated compared to the national baseline of 2.1% of visits for ILI
How has the time period of the pandemic (2019-2022) increased or decreased the percentage of ILI related visits compared to the National Baseline
How has the time period of the pandemic (2019-2022) increased or decreased the percentage of ILI related visits compared to the National Baseline
2019-2020
2019-2020
ILI related visits spike to almost 3 times the National Baseline
2020-2021
2020-2021
ILI visits throughout the entire season are lower than the National Baseline
2021-2022
2021-2022
There is a single spike during winter holidays but after that, it remains below the National Baseline
Does the East North Central division have more cases of influenza than the New England division during 2012-2013?
Does the East North Central division have more cases of influenza than the New England division during 2012-2013?
East North Central division Positive Influenza Cases
East North Central division Positive Influenza Cases
Total positive cases: 4,576
New England division Positive Influenza Cases
New England division Positive Influenza Cases
Total positive cases: 3,471
The East North Central division had ~1,100 more cases than the New England division
The East North Central division had ~1,100 more cases than the New England division
For the Mid-Atlantic?
For the Mid-Atlantic?
Mid-Atlantic division Positive Influenza Cases
Mid-Atlantic division Positive Influenza Cases
Total positive cases: 10,653
The Mid-Atlantic division had more than twice as much positive influenza cases than the East North Central division
The Mid-Atlantic division had more than twice as much positive influenza cases than the East North Central division
When do the most cases of influenza happen every year?
When do the most cases of influenza happen every year?
2003-2004
2003-2004
2011-2012
2011-2012
2015-2016
2015-2016
2017-2018
2017-2018
By choosing a representative sample of four different seasons, we can see there is a trend that between the end of the year and the beginning of the new year there is an increase in the amount of influenza cases.
By choosing a representative sample of four different seasons, we can see there is a trend that between the end of the year and the beginning of the new year there is an increase in the amount of influenza cases.
What works?
What works?
I like the level of constraints that the users are able to control. The user can control the season that the data is showing, for certain years (2015 and beyond) the user is able to choose where the data being visualized comes from, public health labs or clinical labs.
I like the level of constraints that the users are able to control. The user can control the season that the data is showing, for certain years (2015 and beyond) the user is able to choose where the data being visualized comes from, public health labs or clinical labs.
The user can also control what parts of the country to focus on, state level, national, census division, or HHS regions
The user can also control what parts of the country to focus on, state level, national, census division, or HHS regions
Another thing I like is that the user is allowed to lower amount of representations that are plotted in order to see only what they want
Another thing I like is that the user is allowed to lower amount of representations that are plotted in order to see only what they want
Default View
Hovering over different plots lets you have access to more information based on what is hovered over
Hovering over different plots lets you have access to more information based on what is hovered over
Selecting a region on the map changes the visualizations in real time
Selecting a region on the map changes the visualizations in real time
Needs Improvements
Needs Improvements
In my opinion, the time selection is poorly designed. The data visualized shouldn't be restricted to just a single year long season between two separate years. If it was divided by year or had an option to combine all the different years that would be better. In addition, the slider doesn't remain consistent between different seasons. For some it starts at 40 and some it starts at 35 which makes it more confusing than it needs to be.
In my opinion, the time selection is poorly designed. The data visualized shouldn't be restricted to just a single year long season between two separate years. If it was divided by year or had an option to combine all the different years that would be better. In addition, the slider doesn't remain consistent between different seasons. For some it starts at 40 and some it starts at 35 which makes it more confusing than it needs to be.
Another area which needs improvement is the pie chart. Personally, I think that the values and percentages composing each slice should be displayed at all times because the legend is not readable enough. There also needs to be a better way in order to see the strains that are very small in comparison to the other strains, some type of zoom so you can actually see the different strains would be helpful. I also think it's redundant if the map visualization also displays the same pie chart as the pie chart section. There should also be a total amount of positive tests instead of making the user sum it manually.
Another area which needs improvement is the pie chart. Personally, I think that the values and percentages composing each slice should be displayed at all times because the legend is not readable enough. There also needs to be a better way in order to see the strains that are very small in comparison to the other strains, some type of zoom so you can actually see the different strains would be helpful. I also think it's redundant if the map visualization also displays the same pie chart as the pie chart section. There should also be a total amount of positive tests instead of making the user sum it manually.
There needs to be an easier way to let a first time user understand the keywords like ILI represent because as of right now, they have to look in an entirely separate document in the help button to translate it.
There needs to be an easier way to let a first time user understand the keywords like ILI represent because as of right now, they have to look in an entirely separate document in the help button to translate it.
There should be an option to either compare two different years together or a way to visualize how trends continue year after year. As of right now, you can only do it one season at a time. It would help compare and contrast the different years.
There should be an option to either compare two different years together or a way to visualize how trends continue year after year. As of right now, you can only do it one season at a time. It would help compare and contrast the different years.
Page updated
Google Sites
Report abuse