Context
This app was part of a hackathon I participated in - MacHacks 2021.
My team and I ventured to optimize the blood supply in a hospital setting.
This is a very relevant problem in many hospitals for a few reasons:
- Blood products have varying shelf lives, as long as 1 year, and as short as 12 days, depending on which fraction of the blood it is.</br>
- Hospitals are constantly having to balance the intake of new donations with the predicted demand for transfusions in the future. Inefficiency in the inventory management of a hospital can lead to the loss of hundreds of units of blood per year. </br>
Idea
Our teams idea was to create an automated end-to-end predictive modelling system that would be able to inform physicians and hospital staff when they would need blood, how much, and what type of blood they would need.
Approach
I was responsible for the development of the front and backend of the app, including the development of a synthetic dataset that modelled the important parameters of a blood bank database.
Result
The app can be found on my shiny dashboard. The source code is available at my github. You can view the 5-minute video submission created by my team and I on Youtube.
[Under development]
As this was part of a very rapid ~6 hour prototyping period, I would not say the product is finished. The goal was to get an app that worked in with the proposed data in that time. As such, I would still like to add an actual predictive model to the backend to better reflect the full goal for the project.