According to the phase relationship between the ECG signals of multiple leads, the algorithm selects two of the lead signals to perform wavelet transform R wave detection at the same time, and finally compares the outputs of the two channels to obtain the R wave detection result. The algorithm uses LMS adaptive filtering for noise elimination preprocessing to make the detection result more accurate. The MIT-BIH ECG database disease signal was used for testing on DSP, and the average correct detection rate reached 99.7%. Keywords: R wave detection; LMS adaptive filtering; multi-lead; DSP R wave detection is the key to ECG signal detection. It is not only the basis for detecting other ECG wave groups (P wave, T wave), but also can help diagnose diseases such as arrhythmia and abnormal heartbeat. For relatively regular normal ECG signals, automatic detection is relatively simple. Usually, only an appropriate threshold can be selected to obtain satisfactory results through the threshold judgment method. However, the process from ECG signal acquisition to transmission often introduces interference noise, such as electromyographic signals, baseline drift, and power frequency interference. To this end, many scholars have proposed effective algorithms [2] to suppress noise and improve the detection rate of R waves [1][3][4]. Among them, the wavelet transform method is widely used because of its good localization characteristics in the time domain and frequency domain. The usual detection method is divided into two steps: signal preprocessing to achieve signal noise elimination and highlight feature points; R wave detection to determine the starting and ending points of the R wave. The published literature is all for single-lead signal detection. Due to the large individual differences of ECG signals, the single-lead automatic detection method is more likely to have singular point misdetection. The algorithm in this paper chooses to perform wavelet decomposition and detection judgment on the ECG signals of two channels at the same time, and then synthesizes the detection judgment of the two channels to obtain the final result. Based on the multi-lead detection method, the two channels can correct the detection results from different angles, thereby improving the correct detection rate. The TMS320VC5509 DSP selected as the hardware chip is a fast, low-power fixed-point digital signal processor. Its operating frequency can reach 144MHz and it integrates 128K bytes of RAM on the chip. Its low power consumption characteristics meet the requirements of sustainable detection and miniaturization.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore