Mini-project 2 allows you an opportunity to delve into exploring data analytics and visualization in order to move towards your individual projects

Healthcare analytics focuses on offering insights into hospital management, patient records, costs, diagnoses, and more. Within the health informatics field, it covers a broad aspect of the healthcare industry, offering insights on both the macro and micro level. When combined with computing standards and data visualization tools, healthcare analytics helps professionals in the field to operate better by providing real-time information that can support decisions and deliver actionable insights.

For your second mini-project, you will have the opportunity to explore and work through a solution of your own, in an area of healthcare of your choice. Note that, where possible, the area you choose can be utilize in future projects. Once you have chosen an area (ex. Mortality/Mortality Rates) you will explore and explain analytics could solve a critical issue in that area (ex. death certification, death reporting). From there, you will architect a solution, potentially utilizing what has been done in or outside of the field, or something "new." You will pull this all together in the form of a tool or Jupyter notebook. Details are listed below. 

Solution Requirements

Tool:  Must include one of the technologies covered in this unit (ie. ClarityNLP, Tableau). Reach out to a TA if you would like to utilize a tool not covered in this unit or course for this project.

Data:  We provide you with data resources in the course, but you are free to find your own or create your own. If you create your own you will need to provide the steps in how you created it (so that it can be recreated) and explain why you created your data the way that you did (should be inspired by clinical data).

Analysis:  Clean and analyze the data as appropriate, explaining your approaches and decisions on all levels.

Visualization: Decide on a visualization approach and implement it based on your area of choice.

Format:  Format should be appropriate to the solution. (Ex: Jupyter Notebook, Google Colab, Github Repo)

Documentation:  Instructions on how to deploy/use your solution and a demo/tutorial.  

Documentation Structure

Please note that the goal of this project is the deployed tool or solution. The documentation should be seen as less of an academic paper here but more of descriptor on how to use and access your solution, with details on what it does and how it does it.

●  Background and Significance

●  Problem

●  Proposed Solution or Idea (you must share the tools used, etc)

●  Complexity or Effort

●  Tutorial or Demonstration

●  References 

Submission

 Complete your assignment

then save your submission as a PDF. You should submit a single PDF for this assignment. If your assignment involves things (like videos, working prototypes, etc.) that cannot be provided in PDF, you should provide them separately (through OneDrive, Google Drive, Dropbox, etc.) and submit a PDF that links to or otherwise describes how to access that material. You must also include notebook files. You must utilize the APA citation format.

This is an individual assignment. All work you submit should be your own. Make sure to cite any sources you reference, and use quotes and in-line citations to mark any direct quotes. You are limited to 8 pages maximum (excluding references). This is a formal assignment, therefore your submission should reflect this. 

The officially supported tools are:

   ● Tableau

   ● Plotly

   ● R Studio.

   ● ClarityNLP   

        ●   Mini-project 2 allows you an opportunity to delve into exploring data analytics and visualization in order to move towards your individual projects (and some of the Team Projects). As with

mini-project 1, do not overthink this. We want you to be able to develop some insight into these topics in order to inform your choice of direction in your individual projects, this is not intended to produce the greatest analysis of healthcare data in the history of the world in a few weeks. 

●   Consider topics related to healthcare data you would like to explore, which may related to

mini-project 1 or entirely new. In terms of complexity think in terms of the Tableau demonstration dashboard detailing a series of visualizations for Covid 19 mortality vs infection rates.

●   This project is geared towards creating dashboards or small analytics studies. Your deliverable does not have to be a full application, but it may be if you'd like. Just keep it simple, this isn't meant to be an intense project, you will have the opportunity to start building out more application wise in mini-project 3 and then for the individual project.

●   The data sources you will want to use will vary greatly depending on what you would like to do. You may check out @161 for some options provided by instructors as well as fellow students. (Please share your own as well!) You may create your own data for this project for demonstration purposes. For instance, if you would like to create a dashboard that shows cancer type by county in the state of Georgia, this might be very difficult data to locate, so you are free to build some simple data to showcase your work. The only requirement here is that any data you create be reasonable. (Of course if you opt for something more research than visualization oriented, creating your own data would likely not be appropriate. In that case, check out the Synpuf or Synthea data sets for how you might approach that.)

●  Data format can be whatever you require for your project. Most data sets are provided as CSV spreadsheets, but if you would like to pull from FHIR resources or something else as well you are free to. (Though that would be more for full applications, which again is not really the scope envisioned here even if still an option.) 

Project Data Sources 

Synpuf

Synpuf is a synthetic data set from the Center for Medicare and Medicaid Services (CMS), which features reasonably complete synhetic health records spanning 2008-2010. This is a very large and complex dataset, which they have split up into multiple parts in the link above. You are free to use whatever portion of the data is necessary for your project.

Synthea 

Synthea is synthetic data generator that allows you to have control in building out records somewhat similar to Synpuf. The link provided here is for the course resource page, which will take you to the Synthea GitHub repo. Keep in mind that for Mini-project 2, Synthea generates records in FHIR by default. You can change this in the settings to output CSVs, though if you really want to incorporate FHIR into mini-project 2 you may. IF YOU USE SYNTHEA YOU MUST USE A SEED FOR THE SAKE OF REPRODUCTION.

WHO Covid 19 Data 

This is a WHO Covid data set which powers their Dashboard. Please remember that if you use this set, you should not be simply recreating their dashboard.

NY Times Covid 19 Data

This is a more varied Covid set from the NYT.  

Please feel free to share any data sets you have found or created below!

Special OH Presentation Notes (Tableau, Plotly, R) 

Tableau Download and Notes

To download Tableau for Academic use you may follow this link: This version is highly recommended as it contains features the trial version does not, though for limited usage the trial version should be fine. The data used in the presentation can be found at

Plotly Notes

Plotly main site: 

Plotly Chart studio: The presentation refers to two additional notes in chat:

   ●   A link to, which shows some uses of Plotly in their Dash framework. (NOTE: Dash is not required and is likely overkill for mini-project 2. This link is purely for some code examples of how it is used.) 

   ●   The additional library referenced is "cuxfilter".  

Tableau and Plotly Questions

Ximin Mi, the presenter for Tableau and Plotly also invites you to contact her if you have any additional questions working with those tools that are not addressed within the demonstrations, at . Please be conscious of the fact that Ximin is with the GT LIbrary and does not work directly with this course, and is not aware of our specific assignments, please only contact her if it is specifically a question related to working with Tableau or Plotly. 

Students are also invited to participate in the GT Library's Visual Critique Thursday events, which may be found listed at.

R Studio Notes

The R Studio demonstration is presented using R Studio Cloud. To follow along with it, create an account at. While you may download R Studio to local machine, R Studio Cloud provides a simple way to ensure you are working in a reliable environment independent of issues that may occur with specific OSs and Architectures. The demonstration uses the day.csv data set found at day.csv - and the R file IntroToRClass_CS.R which can be downloaded at. If you have any issues with these links, both files may be found in the Files section of Canvas.