False arrhythmia alarm reduction in the intensive care unit False alarms constitute more than 80% of alarms triggered in the intensive care unit. This has severe implications for disruption of patient care and desensitization from clinical staff to true alarms. A method to reduce this high false alarm rate would therefore greatly benefit patients as well as nurses in the ability to provide care. In this SuperUROP project we build upon previous work to build a robust false arrhythmia alarm reduction system. We make use of signal processing and machine learning techniques to determine if channels exhibit evidence for cardiac arrhythmias. We hope to build an algorithm which performs with high sensitivity and specificity in a retrospective and real-time setting. Such an algorithm could be translated for use in the ICU to promote overall patient care.
I am a junior studying computer science at MIT and I find the application of computer science towards solving medical problems to be fascinating. I am very excited for my project which uses signal processing and machine learning to identify false alarms in cardiac arrhythmia detection in the ICU. I hope to learn signal processing techniques and machine learning principles in the various ways to help improve medicine.