Current at-home cardiopulmonary resuscitation (CPR) training systems are limited by the feedback they provide. Virtual trainers have the potential to enhance feedback in CPR training systems by providing real-time demonstrations. We developed CPRBuddy, a CPR training system that uses a virtual trainer to emulate a live trainer to automatically provide powerful feedback. CPRBuddy was evaluated via a user study consisting of 9 participants comparing CPR compressions both with and without feedback from CPRBuddy. We found that CPRBuddy’s demonstrative feedback has potential to improve the immediate performance of CPR, e.g., compression depth, frequency, and recoil. This work contributes towards the design of avatars for training.

Sarah Morrison-Smith, Heng Yao, Isaac Wang, Benjamin Lok, and Jaime Ruiz. 2018. Staying Alive With Virtual Humans. In Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems (CHI EA ’18). Association for Computing Machinery, New York, NY, USA, Paper LBW038, 1–6. https://doi.org/10.1145/3170427.3188477

@inproceedings{10.1145/3170427.3188477,
author = {Morrison-Smith, Sarah and Yao, Heng and Wang, Isaac and Lok, Benjamin and Ruiz, Jaime},
title = {Staying Alive With Virtual Humans},
year = {2018},
isbn = {9781450356213},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3170427.3188477},
doi = {10.1145/3170427.3188477},
abstract = {Current at-home cardiopulmonary resuscitation (CPR) training systems are limited by the feedback they provide. Virtual trainers have the potential to enhance feedback in CPR training systems by providing real-time demonstrations. We developed CPRBuddy, a CPR training system that uses a virtual trainer to emulate a live trainer to automatically provide powerful feedback. CPRBuddy was evaluated via a user study consisting of 9 participants comparing CPR compressions both with and without feedback from CPRBuddy. We found that CPRBuddy's demonstrative feedback has potential to improve the immediate performance of CPR, e.g., compression depth, frequency, and recoil. This work contributes towards the design of avatars for training.},
booktitle = {Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems},
pages = {1–6},
numpages = {6},
keywords = {cardiopulmonary resuscitation (cpr), virtual trainer, multimodal feedback},
location = {Montreal QC, Canada},
series = {CHI EA '18}
}