Machine learning as a supportive tool to recognize cardiac arrest in emergency calls

Abstract

Background: Emergency medical dispatchers fail to identify approximately 25% of cases of out of hospital cardiac arrest, thus lose the opportunity to provide the caller instructions in cardiopulmonary resuscitation. We examined whether a machine learning framework could recognize out-of-hospital cardiac arrest from audio files of calls to the emergency medical dispatch center.Methods: For all incidents responded to by Emergency Medical Dispatch Center Copenhagen in 2014, the associated call was retrieved. A machine learning framework was trained to recognize cardiac arrest from the recorded calls. Sensitivity, specificity, and positive predictive value for recognizing out-of-hospital cardiac arrest were calculated. The performance of the machine learning framework was compared to the actual recognition and time-torecognition of cardiac arrest by medical dispatchers.Results: We examined 108,607 emergency calls, of which 918 (0.8%) were out-of-hospital cardiac arrest calls eligible for analysis. Compared with medical dispatchers, the machine learning framework had a significantly higher sensitivity (72.5% vs. 84.1%, p < 0.001) with lower specificity (98.8% vs. 97.3%, p < 0.001). The machine learning framework had a lower positive predictive value than dispatchers (20.9% vs. 33.0%, p < 0.001). Time-torecognition was significantly shorter for the machine learning framework compared to the dispatchers (median 44 seconds vs. 54 s, p < 0.001).Conclusions: A machine learning framework performed better than emergency medical dispatchers for identifying out-of-hospital cardiac arrest in emergency phone calls. Machine learning may play an important role as a decision support tool for emergency medical dispatchers.

Towards definitions of time-sensitive conditions in prehospital care

“…An alternative approach could be to develop a clinical decision support system with machine learning capabilities for all these symptoms, in order to help EMS clinicians to distinguish patients with time-sensitive conditions from those without. For example, machine learning support tools have shown higher sensitivity and faster identification over the telephone in recognising cardiac arrest compared with professional dispatchers [16]. These instruments need to be tested in large patient cohorts in order to prove their eventual value.…”

Wireless monitoring and artificial intelligence: A bright future in cardiothoracic surgery

“…18 Deep learning has also been used to predict outcomes in heart failure 19 and to identify out-of-hospital cardiac arrest in emergency telephone calls in Denmark. 20 A network model using an AI-enabled electrocardiogram was successful in screening patients for asymptomatic left ventricle dysfunction. 21 The integration of AI technology and biosensors with the wireless capabilities through Bluetooth, Wi-Fi, and global positioning systems have led to the development of pointof-care (POC) diagnosis.…”

Deep multitask ensemble classification of emergency medical call incidents combining multimodal data improves emergency medical dispatch

“…perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 28, 2020. ; https://doi.org/10.1101/2020.06.26.20123216 doi: medRxiv preprint volume, [31] automatic stress detection of the caller, [32] interpretable knowledge extraction, [33] performance monitoring, [34] cardiac arrest calls assistance [35] or triaging unconscious and fainting patients. [36] Therefore, we can argue that deep learning models are a feasible and promising technology to improve EMD through EMCI classification.…”