The American Heart Association states that more than 350,000 cardiac arrests occur outside of hospitals each year, and about 90% of those are fatal. With only 11% of Americans knowing how to correctly perform CPR, it is crucial to provide bystanders with the resources and training to provide high-quality compressions to improve outcomes. Current CPR feedback devices on the market are prohibitively expensive and primarily target medical providers, failing to meet the needs of bystanders without a background in health care. Our app-based CPR device, the CPR Guardian, fills these needs, providing a layperson-targeted CPR guide, tempo guidance, instructional voice commands, 911 notifications, and real-time audio compression depth feedback all contained within the user’s phone, increasing accessibility to CPR guidance due to the pervasiveness of smartphones. The developed depth compression feedback algorithm was validated using data collected from multiple subjects performing CPR on a mannequin. Two-sample unequal variance student’s t-tests, Kolmogorov-Smirnov tests, and Hellinger Distance tests were performed to assess the relationship between our app-based depth compression feedback algorithm, data filtered manually in MATLAB and JS, and feedback from the PRESTAN Professional Adult Mannequin. Results show that an excellent match is achieved between the depth values returned by the algorithm embedded in the mobile application and those computed by established feedback methods. Additionally, preliminary data revealed that the quality of CPR performed by untrained laypersons improved with app usage. These results reflect the potential for this novel development to reduce out-of-hospital cardiac deaths.

The group would like to thank their faculty sponsors Dr. Gary Bledose and Dr. Roobik Gharabagi for their support of this project.