Real-time ECG analysis using a TI TMSC54/spl times/ digital signal processing chip

Real-time template-based correlation was implemented on the Texas Instruments TMSC5402 digital signal processing (DSP) chip to allow for the detection of abnormal QRS-complexes in electrocardiographic signals. Several novel techniques were used to achieve real-time implementation of correlation algorithms, as well as development of an automatic template generation module. The main goal was the successful development of an algorithm to rapidly and efficiently detect irregular QRS complexes. This is implemented by software embedded in a DSP chip using limited memory resources. Once the algorithm produces a favorable correlation of a normal QRS, the timer is reset (and alarm deactivated, if active) indicating a normal passage. Values of the timer and correlation threshold are also displayed on an oscilloscope to allow for visual inspection of the system while executing. This paper describes an academic undergraduate semester-long project in the implementation of an arrhythmia (abnormal beat) detector accomplished via a standard DSP chip. The bulk of the project consisted of software design, which was kept modular to facilitate the visualization of project flow and modification of individual software components. The system can easily be implemented on a small DSP, field programmable gate array (FPGA), or an application specific integrated circuit (ASIC). The device can be bundled with a micro hard-drive to record ECG signals over a lengthy period of time in minimal space.

1 Introduction

Electrocardiography is a method to record the electrical activity of the heart. It is used to measure the rate and regularity of heartbeats. ECG analysis can also indicate the size and function of the heart chambers, any damage to the heart, and the effects of drugs. By examining the EeG signal, we can tell whether a ventricular depolarization results in a heartbeat that is normal. Figure 1.

schematic ecg signal showing the characteristic waveforms, intervals, and segments.