45 - Collection of Health Data from Smart Devices

Description/Abstract/Artist Statement

Collection of Health Data from Smart Devices

Cameron Zell, Peter Scheible, He Jing

Within the field of health informatics, we have seen the widespread adoption of wearable devices such as watches and continuous glucose monitors (e.g. Dexcom) that record and track important metabolic health data such as heartrate, blood glucose, and physical activity levels. These devices have enabled the collection of valuable health data that can be used for predictive health monitoring and decision-making, particularly useful within machine learning models. However, key issues lie in the fragmented nature of these data sources - the inconsistency in various data formats, and the lack of a standardized method to retrieve and centralize a large amount of this metabolic health data efficiently. The overall goal of the project is to address this challenge by exploring the integration of health data across multiple platforms (such as Dexcom and Fitbit) and providing a method by which we can harvest and centralize these datasets from participants for future use in predictive models. Datasets such as these are essential in the development of more robust machine learning methods that will yield more accurate and timely predictions for individuals with metabolic diseases, enabling more personalized, effective, and responsive health interventions.

Presenting Author Name/s

Cameron Zell

Faculty Advisor/Mentor

He Jing

Faculty Advisor/Mentor Department

SCI Computer Sciences

College Affiliation

College of Sciences

Presentation Type

Poster

Disciplines

Artificial Intelligence and Robotics | Databases and Information Systems

This document is currently not available here.

Share

COinS
 

45 - Collection of Health Data from Smart Devices

Collection of Health Data from Smart Devices

Cameron Zell, Peter Scheible, He Jing

Within the field of health informatics, we have seen the widespread adoption of wearable devices such as watches and continuous glucose monitors (e.g. Dexcom) that record and track important metabolic health data such as heartrate, blood glucose, and physical activity levels. These devices have enabled the collection of valuable health data that can be used for predictive health monitoring and decision-making, particularly useful within machine learning models. However, key issues lie in the fragmented nature of these data sources - the inconsistency in various data formats, and the lack of a standardized method to retrieve and centralize a large amount of this metabolic health data efficiently. The overall goal of the project is to address this challenge by exploring the integration of health data across multiple platforms (such as Dexcom and Fitbit) and providing a method by which we can harvest and centralize these datasets from participants for future use in predictive models. Datasets such as these are essential in the development of more robust machine learning methods that will yield more accurate and timely predictions for individuals with metabolic diseases, enabling more personalized, effective, and responsive health interventions.