Abstract
Diabetes management is one of the most challenging aspects of improving population health. New technology has allowed healthcare providers to track their patients’ blood glucose levels in real-time, but current tools have poorly organized the influx of patient data for providers to draw any useful and actionable insights. In the past year, Penn Medicine obtained data from patient-entered flowsheets, which led to questions on how to use, interpret, aggregate, and present this data to providers.
Our team has developed a web dashboard to determine and display patients’ monthly blood glucose levels in a more useful way for Penn Medicine providers to make immediate clinical decisions. Our dashboard distinguishes itself from existing diabetes management tools by autonomously classifying blood glucose readings into one of four different categories: Fasting, Preprandial, Postprandial, and Random. The two necessary inputs for this classification process are patient blood glucose data and patient-entered mealtime data.
Throughout the year, we have met with health providers at Penn Medicine to discuss what would be the best way to view the influx of patient data, along with ways to update the web app.
The web dashboard’s friendly user interface enables health providers to view and understand both Population and Patient Level Summaries of their diabetic patients. The software code, which includes blood glucose data processing and manipulation, is written in HTML/CSS/JavaScript and utilizes the D3.js library for graphical plots.
Link to PowerPoint Slides:
https://docs.google.com/presentation/d/1dbFPpjXgLynqzMLhcnb79C422Lc1lKxvWXvV1h6eoqA/edit#slide=id.p
Log in or sign up for Devpost to join the conversation.