I-Corps: Artificial Intelligence-based mobile application to mitigate health risks of firefighters

New York University, New York NY

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is the development of an Artificial Intelligence (AI)-based mobile application to reduce firefighter deaths. Firefighters are fourteen times more likely to suffer a sudden cardiac event (SCE) in response to an alarm, and 136 times more likely to suffer a fatal SCE after firefighting than nonemergency duties. The proposed technology is designed to monitor the physiological data of firefighters using commercially available personal wireless wearable health trackers, evaluate cardiovascular risk factors, and alert the firefighter about impending risks. The goal for this proposed technology is to reduce firefighter mortalities and morbidities associated with cardiovascular disease. Approximately 92% of out-of-hospital cardiac arrests result in death suggesting that many people do not recognize the symptoms, and don’t act on early warning signs. Early identification of cardiovascular health risks (e.g., hypertension, arrhythmia, sleep apnea) that increase the possibility of SCE, continuous monitoring of disease progression, and addressing risks in a timely manner may increase the chances of preventing or surviving a SCE. The health trackers are low-cost, and can continuously track physiological data (heart rate, sleep behavior) and physical activity (step count, travel distance, floors climbed), and may be used to upload this data wirelessly to the cloud via a mobile device. Long-term benefits may include a decrease in cardiac incidence that may lead to lower healthcare costs and improve the safety of firefighters. In addition, the proposed technology may be used by the general population, which may have a broad health benefit to society. This I-Corps project is based on the development of a mobile software application for the fire service to reduce firefighter deaths due to sudden cardiac events (SCE). The proposed technology utilizes commercially available health tracker technology integrated with Artificial Intelligence (AI) models that have been trained and developed using firefighters’ health data and customized using relevant clinical studies from the fire service. The proposed software is built with react-native cross-platform (iOS/Android) and is integrated with deep learning models to remotely diagnose arrhythmia, hypertension, atrial fibrillation, and sleep apnea, which are prevalent cardiovascular risk factors in fire service data leading to sudden cardiac events – a leading cause (50-60%) of firefighter morbidity and on-duty deaths. The proposed Internet of Things system captures the physiological data from wearable health trackers and feeds it to cloud-based AI models for real-time remote diagnosis. Based on custom risk thresholds, the user may then be notified when a health risk is detected by the AI models. In addition, wide-spread prolonged usage of the proposed software application also may provide a database of firefighters’ physiological data that may then be used to further refine the technology, lead to new understandings of physiological, and potentially pathological, responses to firefighting, and stimulate new data-driven clinical research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.