A recent study published in Fire Safety Journal examines how machine learning, which is a type of artificial intelligence, is being used to recognize abnormal heart rhythms, also known as heart arrhythmia, in on-duty firefighters, which resulted in the deaths of 36 on-duty firefighters in 2022, per data from the National Fire Protection Association. This study was conducted by researchers at the National Institute of Standards and Technology (NIST) and holds the potential to develop a portable heart monitor device that on-duty firefighters could put on while conducting emergency calls. This device would notify them of irregular heart rhythms and allow them to seek appropriate medical attention before their condition worsens.
“Year after year, sudden cardiac events are by far the number one killer of firefighters,” said Dr. Chris Brown, who is a researcher at NIST, and a co-author on the study. “Cardiac events also cause career-ending injuries and long-term disabilities.”
For the study, the researchers trained a machine learning algorithm to develop a model with the goal of being able to detect irregular heart rhythms in on-duty firefighters. Using 10-year-old electrocardiogram (ECG) data collected at the University of Rochester from 112 firefighters for a period of 24 hours each. This included on-duty shifts comprised of 16 hours and off-duty shifts of eight hours. This data was then used by the researchers to develop what they refer to as the Heart Health Monitoring (H2M) model, which is designed to detect normal and irregular heart rhythms, including ventricular tachycardia or atrial fibrillation.
“The model is designed to effectively learn ECG patterns from both normal and abnormal beats,” said Dr. Jiajia Li, who is a guest researcher at NIST and lead author of the study.
After the model was developed and trained, it was then used to detect ECG data not observed in the previous Rochester dataset, including 6,000 irregular ECG samples, and recognized them with an accuracy rate of 97 percent. Additionally, the researchers conducted what’s known as a parametric study where they used H2M to analyze non-firefighter ECG datasets, which demonstrated an error rate of approximately 40 percent.
Going forward, the researchers hope to use this model to develop portable heart monitors that that on-duty firefighters could put on while conducting emergency calls, which would notify them of irregular heart rhythms and allow them to seek appropriate medical attention before their condition worsens.
“This technology can save lives,” said Dr. Wai Cheong Tam, who is a researcher at NIST and a co-author on the study. “It could benefit not only firefighters but other first responders and additional populations in the general public.” Dr. Tam noted this method could be expanded to assist other groups if the AI is provided with suitable ECG datasets.
How else will artificial intelligence be used to potentially save lives in the coming years and decades? Only time will tell, and this is why we science!
Sources: Fire Safety Journal, Mayo Clinic, National Fire Protection Association, EurekAlert!, Arm Limited, ScienceDirect
As always, keep doing science & keep looking up!