Classification of Distress Alarms in Intensive Care Unit Using Neural Network

Intensive care units (ICUs) are dedicated to the treatment of patients with severe or life-threatening illnesses and injuries, necessitating constant care, close supervision, and the use of life support equipment and medications to maintain normal bodily functions. In ICUs, the monitoring of patients' physiological parameters has significantly advanced patient safety by alerting ICU staff through visual or audible alarms. However, due to understaffing during night shifts, nurses may not be able to respond promptly to these alarms, potentially endangering patient lives. Moreover, the continuous exposure of medical staff to alarms may lead to decreased concentration and errors, as well as desensitization to critical alarms, further compromising patient safety. The proposed solution utilizes a neural network to classify distress alarms in the ICU and direct them to a nurse station for timely response.