Machine learning as a supportive tool to recognize cardiac arrest in emergency calls. Blomberg, S. N., Folke, F., Ersbøll, A. K., Christensen, H. C., Torp-Pedersen, C., Sayre, M. R., Counts, C. R., & Lippert, F. K. Resuscitation, 138:322–329, May, 2019. doi abstract bibtex 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-to-recognition 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 \textless 0.001) with lower specificity (98.8% vs. 97.3%, p \textless 0.001). The machine learning framework had a lower positive predictive value than dispatchers (20.9% vs. 33.0%, p \textless 0.001). Time-to-recognition was significantly shorter for the machine learning framework compared to the dispatchers (median 44 seconds vs. 54 s, p \textless 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.
@article{blomberg_machine_2019,
title = {Machine learning as a supportive tool to recognize cardiac arrest in emergency calls},
volume = {138},
issn = {1873-1570},
doi = {10.1016/j.resuscitation.2019.01.015},
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-to-recognition 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 {\textless} 0.001) with lower specificity (98.8\% vs. 97.3\%, p {\textless} 0.001). The machine learning framework had a lower positive predictive value than dispatchers (20.9\% vs. 33.0\%, p {\textless} 0.001). Time-to-recognition was significantly shorter for the machine learning framework compared to the dispatchers (median 44 seconds vs. 54 s, p {\textless} 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.},
language = {eng},
journal = {Resuscitation},
author = {Blomberg, Stig Nikolaj and Folke, Fredrik and Ersbøll, Annette Kjær and Christensen, Helle Collatz and Torp-Pedersen, Christian and Sayre, Michael R. and Counts, Catherine R. and Lippert, Freddy K.},
month = may,
year = {2019},
pmid = {30664917},
keywords = {Artificial intelligence, Cardiopulmonary resuscitation, Detection time, Dispatch-assisted cardiopulmonary resuscitation, Emergency medical services, Machine learning, Out-of-hospital cardiac arrest},
pages = {322--329}
}
Downloads: 0
{"_id":"q25GAnnpzTimpcWDQ","bibbaseid":"blomberg-folke-ersbll-christensen-torppedersen-sayre-counts-lippert-machinelearningasasupportivetooltorecognizecardiacarrestinemergencycalls-2019","authorIDs":[],"author_short":["Blomberg, S. N.","Folke, F.","Ersbøll, A. K.","Christensen, H. C.","Torp-Pedersen, C.","Sayre, M. R.","Counts, C. R.","Lippert, F. K."],"bibdata":{"bibtype":"article","type":"article","title":"Machine learning as a supportive tool to recognize cardiac arrest in emergency calls","volume":"138","issn":"1873-1570","doi":"10.1016/j.resuscitation.2019.01.015","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-to-recognition 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 \\textless 0.001) with lower specificity (98.8% vs. 97.3%, p \\textless 0.001). The machine learning framework had a lower positive predictive value than dispatchers (20.9% vs. 33.0%, p \\textless 0.001). Time-to-recognition was significantly shorter for the machine learning framework compared to the dispatchers (median 44 seconds vs. 54 s, p \\textless 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.","language":"eng","journal":"Resuscitation","author":[{"propositions":[],"lastnames":["Blomberg"],"firstnames":["Stig","Nikolaj"],"suffixes":[]},{"propositions":[],"lastnames":["Folke"],"firstnames":["Fredrik"],"suffixes":[]},{"propositions":[],"lastnames":["Ersbøll"],"firstnames":["Annette","Kjær"],"suffixes":[]},{"propositions":[],"lastnames":["Christensen"],"firstnames":["Helle","Collatz"],"suffixes":[]},{"propositions":[],"lastnames":["Torp-Pedersen"],"firstnames":["Christian"],"suffixes":[]},{"propositions":[],"lastnames":["Sayre"],"firstnames":["Michael","R."],"suffixes":[]},{"propositions":[],"lastnames":["Counts"],"firstnames":["Catherine","R."],"suffixes":[]},{"propositions":[],"lastnames":["Lippert"],"firstnames":["Freddy","K."],"suffixes":[]}],"month":"May","year":"2019","pmid":"30664917","keywords":"Artificial intelligence, Cardiopulmonary resuscitation, Detection time, Dispatch-assisted cardiopulmonary resuscitation, Emergency medical services, Machine learning, Out-of-hospital cardiac arrest","pages":"322–329","bibtex":"@article{blomberg_machine_2019,\n\ttitle = {Machine learning as a supportive tool to recognize cardiac arrest in emergency calls},\n\tvolume = {138},\n\tissn = {1873-1570},\n\tdoi = {10.1016/j.resuscitation.2019.01.015},\n\tabstract = {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.\nMETHODS: 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-to-recognition of cardiac arrest by medical dispatchers.\nRESULTS: 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 {\\textless} 0.001) with lower specificity (98.8\\% vs. 97.3\\%, p {\\textless} 0.001). The machine learning framework had a lower positive predictive value than dispatchers (20.9\\% vs. 33.0\\%, p {\\textless} 0.001). Time-to-recognition was significantly shorter for the machine learning framework compared to the dispatchers (median 44 seconds vs. 54 s, p {\\textless} 0.001).\nCONCLUSIONS: 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.},\n\tlanguage = {eng},\n\tjournal = {Resuscitation},\n\tauthor = {Blomberg, Stig Nikolaj and Folke, Fredrik and Ersbøll, Annette Kjær and Christensen, Helle Collatz and Torp-Pedersen, Christian and Sayre, Michael R. and Counts, Catherine R. and Lippert, Freddy K.},\n\tmonth = may,\n\tyear = {2019},\n\tpmid = {30664917},\n\tkeywords = {Artificial intelligence, Cardiopulmonary resuscitation, Detection time, Dispatch-assisted cardiopulmonary resuscitation, Emergency medical services, Machine learning, Out-of-hospital cardiac arrest},\n\tpages = {322--329}\n}\n\n","author_short":["Blomberg, S. N.","Folke, F.","Ersbøll, A. K.","Christensen, H. C.","Torp-Pedersen, C.","Sayre, M. R.","Counts, C. R.","Lippert, F. K."],"key":"blomberg_machine_2019","id":"blomberg_machine_2019","bibbaseid":"blomberg-folke-ersbll-christensen-torppedersen-sayre-counts-lippert-machinelearningasasupportivetooltorecognizecardiacarrestinemergencycalls-2019","role":"author","urls":{},"keyword":["Artificial intelligence","Cardiopulmonary resuscitation","Detection time","Dispatch-assisted cardiopulmonary resuscitation","Emergency medical services","Machine learning","Out-of-hospital cardiac arrest"],"downloads":0},"bibtype":"article","biburl":"https://api.zotero.org/users/2544467/collections/T7ZNR5XD/items?key=fEo577bQhYtG6H1aJTHD5odr&format=bibtex&limit=100","creationDate":"2019-11-10T14:49:16.191Z","downloads":0,"keywords":["artificial intelligence","cardiopulmonary resuscitation","detection time","dispatch-assisted cardiopulmonary resuscitation","emergency medical services","machine learning","out-of-hospital cardiac arrest"],"search_terms":["machine","learning","supportive","tool","recognize","cardiac","arrest","emergency","calls","blomberg","folke","ersbøll","christensen","torp-pedersen","sayre","counts","lippert"],"title":"Machine learning as a supportive tool to recognize cardiac arrest in emergency calls","year":2019,"dataSources":["dbq9tgqw6Eg4GQjKB"]}