Assignment 4. Designing a data science health app


Over a million new life science publications come out every year. Even medical specialists have a hard time keeping up with their own fields. Further, new data sources highly relevant to patients and doctors are appearing constantly, such as physician reported adverse events, and side effects reported on social media.

How can patients, along with their physicians, find the data most relevant to them and their conditions, as a starting point for discussion? Increasingly, people have multiple chronic conditions (diabetes, cardiovascular disease, cancer, obesity) as well as various other conditions specific to them. In this assignment, you will design a smartphone/tablet app (iPhone, iPad, Android) to help people find the most pressing information relevant to them and their conditions, including latest research findings and adverse event reports. Note that you do not need to implement the app, but to make a design for one. There is not one right answer here - there are many kinds of apps that could help patients. You can thus be very creative. Again, your aim here is to think about the design of the app, so do don't be constrained by whether or not you think it can be easily implemented using current technology.

Think about you or people you know that have been through one or more complex health conditions (diabetes, cardiovascular disease, cancer, etc). What would have been helpful? Here is one "Patient X" you might use as an example as you design your app:

Name: Patient X
Age: 57
Sex: Male
Weight: 350lbs
Medical history: Cardiovascular patient, 1 previous heart attack 5 years ago, pre-diabetes
Reported side effects: Muscle cramps, migrane headaches
Drugs taken:
- Fenofibrate - PPAR-α activator which lowers LDL/VLDL, triglycerides, increases HDL.
- Levothyroxine - Synthetic thyroid hormone used to treat hypothyroidism
- Valsartan/Hydrochlorothiazide - Valsartan is an angiotensin II antagonist for hypertension
- Aspirin - as a blood thinner
- Rosuvastatin (Crestor) - HMG-CoA inhibitor, LDL lowering drug

Here are some questions that Patient X and his physician may have:
  • Is the set of drugs I'm taking the most effective one for my condition?
  • Will any of these drugs make me at higher risk of getting diabetes?
  • Are my headaches and cramps related to any of the drugs I'm taking, and if so, can we do anything about it?
  • What's the latest scientific research on people with a similar medical history?

Here are some of the sources of data your might consider using if you want to make your app prototype more realistic (you can try downloading some relevant data for your prototype - but you don't have to do this):
Other sources that might be good for the app, but which you won't have access to for this prototype (although you can "make it up"):
  • Electronic medical records data
  • Structured data extracted from social media

Once you have though through a basic design, implement it with a prototyping app like InVision Prototyping App, AppCooker, or a similar tool.

Submit on Canvas a PDF containing screenshots from your app prototype, showing all functions, and a short (1-2 page) description of what your app does and why. Remember this is a data science course, so be sure to explain how you would analyze and process data in the app.