The advancement of the internet to the paradigm of the Internetof Things (IoT) has brought to society new ways of generating,sharing and using information. The evolution of computing capacityand energy savings in IoT equipament combined with bettersoftware can enabled several new applications, among which wecan highlight the monitoring of people’s health through pervasivedevices connected to the body. In view of this, this work proposesan algorithm to detect atypical situations such as falls in the elderlyand other groups that need health care using accelerometerscontained in wearable devices, particularly smartwatches. For theexperimental evaluation of the proposed algorithm, a database thatcontains data from wearable sensors, environmental sensors, andvisual devices was employed. The metrics used in the evaluationwere accuracy, precision, recall and f1-score, with recall being themost relevant metric in the context. Results show that the bestconfiguration of the algorithm is able to identify falls with 96%recall and F1-score of 90%.