Introduction

The workshop is divided into four parts:

  1. In the first part, we create a SageMaker notebook to generate dummy patient population using Synthea. We then create a HealthLake Data Store, import the data generated using Synthea and then export it to an S3 bucket.
  2. In the second part, we transform the exported data using Glue to extract the information added by the HealthLake. For example, the Glue Job extracts the Extension field JSON and puts it into a separate table that can be used for building the dashboard.
  3. The third part uses Athena to join the individual tables into higher level combined tables.
  4. The fourth part gives instructions on how to build the QuickSight dashboard.

After completing this workshop you will be ready to start generating insights from your organization’s data.

The workshop should take around 2 hours to complete from start to end including the QuickSight dashboard.

Before starting the workshop, download the two files shown below. We will need them in Step 1 and Step 2. (Right click -> Save as -> Save to a folder on your machine)

Cloudformation Template for Step 1

Cloudformation Template for Step 2

Cost

The following are the costs associated with this workshop

  1. S3 bucket storage - The cost for S3 storage will be <$1
  2. HealthLake cost - The cost for HealthLake for the duration of this workshop will be <$5. Please note that HealthLake will continue incurring charges unless you delete Data Store after the workshop.
  3. SageMaker notebook cost - The cost for the SageMaker notebook will be less than <$1 for the duration of the workshop.
  4. Glue Charge - The cost for the Glue job will be <$2 for the duration of the workshop.
  5. Athena Charge - The cost for Athena will be <$2 for the duration of the workshop.

The overall cost for the workshop will be <$15

Clean up

Please ensure that you clean up resources so that your account is not charged after the workshop.

  1. S3 bucket - The CloudFormation in Step 1 creates an S3 bucket. The first step in cleanup is to empty the S3 bucket. Follow the steps here to empty the bucket. The name of the bucket can be found from the output tab of the CloudFormation stack.

  2. QuickSight subscription - As part of this workshop, we will be creating a QuickSight subcription. If you have an existing subscription then you can delete the dataset, analysis and dashboard that you create in this workshop. If you have created a subscription for this workshop, then you can keep the subscription or unsubscribe using the steps shown here - https://docs.aws.amazon.com/quicksight/latest/user/closing-account.html

  3. HealthLake Data Store. - To delete the datastore follow these steps: a. Open AWS Console, and search for HealthLake. b. Open HealthLake console, click on ‘Data Stores’ link on the left. c. Select the checkbox in the data store that says ‘healthlake-workshop’ and click on ‘Delete’.

  4. CloudFormation stacks - Delete the CloudFormation stack created in step 2 and then the stack created in step 1. The steps to delete the stack can be found here