Skip to content

Latest commit

 

History

History
33 lines (17 loc) · 1.53 KB

DIABETES-CHALLENGES.md

File metadata and controls

33 lines (17 loc) · 1.53 KB

Diabetes Queensland

Challenge 1: Type 1 Diabetes Diagnosis

Diagnosis of Type 1 Diabetes is often delayed. Because it is such a small percentage of the total population, it is not top-of-mind for General Practitioners.

How could technology be used to empower either General Practitioners or the General Public to diagnose Type 1 Diabetes when it presents?

Challenge 2: Type 2 Diabetes Prevalence

Are there patterns in the occurrence of Type 2 Diabetes by location? You might like to correlate occurrence with other datasets to see if you can make assertions about risk factors based on data.

IoT Challenge for Diabetes

Many Type 1 Diabetics now have Continuous Glucose Monitors that produce data about the person's Blood Glucose Level - up to 280 data points per day.

There is a an open source project "CGM in the Cloud" that has a "Deploy to Azure" and "Deploy to Heroku" button on it.

There is no "Deploy to AWS" button.

Here are three challenges around these IoT devices:

  • Create a "Deploy to AWS" feature

    • At the moment users must configure a MongoDB on another provider to deploy CGM in the Cloud - can it be done as a one-stop shop with AWS?
    • Can it be deployed in the free tier of AWS?

  • Create an opt-in "citizen data collection"

    • Allows users to choose to contribute to a Big Data dataset of anonymized aggregated data that can then be used by researchers and for machine learning.

  • Can you add machine learning so that the program gives recommendations to users?