Byrsell F, Claesson A, Ringh M, et al. - Researchers conducted an observational study of emergency calls in order to determine if a machine learning framework (ML) can aid in increasing the proportion of calls identifying out-of-hospital cardiac arrest (OHCA) within the first minute compared with dispatchers. In addition, they sought to present the performance of ML with different false positive rate (FPR) settings, as well as examined call characteristics influencing OHCA recognition. Configuration of ML is possible with different FPR settings, ie, more or less inclined to suspect an OHCA depending on the predefined setting. Of 851 OHCA calls, the ML recognized 36% (n = 305) within 1 minute vs 25% (n = 213) recognized by dispatchers. Given the recognition of a higher proportion of OHCA by ML within the first minute compared with dispatchers, ML is suggested to have the potential to be a supportive tool during emergency calls.