Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Systems and Biomedical Engineering

Sudden cardiac arrest (SCA) is a leading cause of death worldwide. It is estimated that more than 3 million people die yearly. Early defibrillation using the automated external defibrillator (AED) has a key role in restoring normal heartbeat of a cardiac arrest patient. AED malfunction can convert the device from a life-saving to a life threatening device. According to FDA, automated external defibrillators (AEDs) are classified as Class-III high-risk medical devices. According to the FDA adverse event database, batteries are the most failure-prone AED components. Good estimation of AED battery life can help avoid AED adverse events, improve AED device reliability, and determine the minimum acceptable warranty period. Having a good battery management plan including a reasonable replacement time and a sufficient warranty period can easily lead to determine safely when the battery end-of-life is. Towards this end, we have investigated AED failure records reported to the FDA which obtained from the Manufacturer and User Facility Device Experience (MAUDE) database with the objective of finding out the best probability model to fit the data. Our results show that the generalized extreme-value probability distribution is the best-fit for the AED battery failure patterns

Issued also as CD