Development and Evaluation of Multimodal Universal CPR AI Assistance and Response Engine (U-CARE)
4 affiliations
- 1Stanford University Online High School, Redwood City, California, USA
- 2Research Fellow, GSE, Stanford University, Stanford, California, USA
- 3Center for Precision Medicine, University of California Davis Health, Davis, California, USA
- 4Department of Anesthesiology, Cleveland Clinic, Cleveland, Ohio, USA
Abstract
Background: Each year, 350,000 people in the United States experience out-of-hospital cardiac arrest, and 90% do not survive. Early cardiopulmonary resuscitation (CPR) can double or triple survival rates.
Project Rationale: Bystander CPR is performed correctly in fewer than 40% of cases, with major challenges persisting: lack of widespread training, decision paralysis, and recall bias during an emergency.
Project Summary: Universal CPR Assistance and Response Engine (U-CARE) is a low-cost smartphone-based system that serves as a universal CPR guide to address these challenges. A large language model–powered verbal guidance system provides objective next-best steps multilingually, coupled with a computer vision model that continuously assesses CPR technique. U-CARE was evaluated using holistic large language model evaluation and computer vision metrics, and outperformed out-of-the-box frontier models on relevance and safety in human evaluation, demonstrating its viability for real-time feedback.
Take-Home Message: U-CARE demonstrates the feasibility of a smartphone-based multimodal CPR feedback and response agent for emergency and training guidance.
Related
- Intelligent evaluation method for inflammatory response after trans-PRK and FS-LASIK based on multimodal feature fusion
- Addressing Measurement Error of Machine-Learned Physical Activity in Nonlinear Dose-Response Survival Analysis: Development and Evaluation of Accelerated Failure Time, Spline, and Simulation-Extrapolation Method
- Explainable Multimodal AI for In-Hospital Detection of Lung Cancer and Heart Failure Using Routine Data: Development, Internal Validation, and Simulated Federated Evaluation
- Development and Exploratory Evaluation of a Habitual Caffeine Intake-Based Index to Characterize the Ergogenic Response to Caffeine in Wrestlers
- Multimodal prediction of neoadjuvant chemotherapy response in locally advanced gastric cancer: a multicenter cohort study
BibTeX
@article{atreja2026development,
title = {{Development and Evaluation of Multimodal Universal CPR AI Assistance and Response Engine (U-CARE)}},
author = {Aarit Atreja and Utkarsh Contractor and Dharan Jaisankar and Piyush Mathur},
journal = {JACC Case Reports},
year = {2026},
month = sep,
doi = {10.1016/j.jaccas.2026.110026},
url = {https://doi.org/10.1016/j.jaccas.2026.110026},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Development and Evaluation of Multimodal Universal CPR AI Assistance and Response Engine (U-CARE)
AU - Aarit Atreja
AU - Utkarsh Contractor
AU - Dharan Jaisankar
AU - Piyush Mathur
PY - 2026
DA - 2026/09/16
JO - JACC Case Reports
DO - 10.1016/j.jaccas.2026.110026
UR - https://doi.org/10.1016/j.jaccas.2026.110026
AB - Background: Each year, 350,000 people in the United States experience out-of-hospital cardiac arrest, and 90% do not survive. Early cardiopulmonary resuscitation (CPR) can double or triple survival rates. Project Rationale: Bystander CPR is performed correctly in fewer than 40% of cases, with major challenges persisting: lack of widespread training, decision paralysis, and recall bias during an emergency. Project Summary: Universal CPR Assistance and Response Engine (U-CARE) is a low-cost smartphone-based system that serves as a universal CPR guide to address these challenges. A large language model–powered verbal guidance system provides objective next-best steps multilingually, coupled with a computer vision model that continuously assesses CPR technique. U-CARE was evaluated using holistic large language model evaluation and computer vision metrics, and outperformed out-of-the-box frontier models on relevance and safety in human evaluation, demonstrating its viability for real-time feedback. Take-Home Message: U-CARE demonstrates the feasibility of a smartphone-based multimodal CPR feedback and response agent for emergency and training guidance.
SN - 2666-0849
ER -