A J2ME Mobile Application for Normal and Abnormal ECG Rhythm Analysis

Introduction

Recently, the patients suffering from cardiovascular disease (CVD) have been undergoing a rapid increase world widely due to the lifestyle change and the aging of population. For many nations such as USA, Australia, European nations, Canada and China, various CVDs are the number one killer while the cerebrovascular diseases (CBD) such as stroke are the number two killer (Roberts, 2006). The prevalence of CVD has risen by 18% over the last decade and is expected to continue to rise over the coming decades due to an expected increase in the elderly population (AIHW, 2004a). In Australia, 37% of the total death in 2001 was caused by cardiovascular diseases and CVD affected total 3.67 million Australian people which is about 18% of national population (AIHW, 2004b). Among Australians having heart attacks, about 25% die within an hour of their first-ever symptoms and over 40% will be dead within a year (Access Economics Report, 2005). CVD together with CBD are also the leading causes of long term disability in adults (Access Economics Report, 2005). They impose a big burden on patients’ families as well as the national healthcare system due to the high costs of care, the resulted lower quality of life, and the premature death. Chronic CVD patients are at high risk of having heart attacks and the majority of such heart attacks take place in out-of-hospital environment where the emergence services cannot be available immediately. Therefore, there is an urgent need to develop a personal monitoring and alarming system which can effectively detect early indications of a heart attack and issue timely warning signals for calling for rescue efforts.

Since last decade, we have witnessed the explosive expansion of the use of mobile phone. Now, mobile phone is one of the most pervasively used single electronics devices in the world. For example, Australia has more subscribed mobile phone handsets than its total population (ACMA Report, 2009). The ever increasing computation power of a mobile phone plus its great mobility make it an ideal pervasive computing platform for telehealth monitoring. Although handheld devices such as mobile phone have been widely proposed to use in various telemedicine applications, they are generally utilized as the wireless data transmission tools. The power from the mobile computing devices has not been fully harnessed. On the other hand, many ambulatory and medical monitoring systems need the acquired vital physiological data be processed in real-time so as to generate the much needed precaution and alarming signals. For such applications, the acquired physiological data should be processed locally, rather than sent to a remote server via a GPRS or 3G mobile telephony network systems, to avoid transmission delay and reduce transmission cost. In order to develop the local intelligence, some limitations pertinent to handheld computing devices need to be addressed. Those limitations include the bottleneck of the bandwidth for large amount of stream data continuous transmission, the partial support of the full extended ASCII set, the limited hardware resources such as processor speed and memory amount, and the restricted programming environment such as no directly support of floating point and no multi-dimensional array support for many mobile phone handsets (Sufi et al., 2006).