Study Of ECG Analysis System

Cardiovascular disease is one of disease which is thratening human being's life. Electrocardiogram (ECG) checking is one of the important ways of diagnosing Cardiovascular disease. Exact ECG automatic analyzing and diagnosing plays an important role in diagnosing Cardiovascular disease. Following with the development of advanced theory, computer technique digital signal process technique and artificial intelligence etal, research of ECG computer automatic analyzing technique continues developing in alternate traversing direction. ECG Automatic analyzing is used in Dynamic Electrocardiogram (DCG). The load which large datas cause is rapidly reduced with the help of automatic analyzing 24 hours'ECG data which recorded by computer.A research model based on others fruit is proposed in this paper and it can integrate experience of ECG specialists and the technologies of engineering. However, the model involved too much work, so a research frame which can be extended dynamic is used in practice. The work preformed at present focused on the utilization of the experience of specialists, and some data analyses method in engineering are also utilized in this work.The following work has been performed in this research:①Introduce a way of filtering ECG signals based on lifting scheme wavelet. There are three interferences which appeared frequently, including myoelectric interference, baseline wamder, frequency interference and it's harmonic component. First, the method destruct original ECG signals into approach signals and delta signals which has different frequency range with lifting scheme wavelet. Second, deal the delta signal with proper threshold by the characteristic of ECG signals. Finally, reconstruct the signals with the reverse of lifting scheme wavelet and erase the main noise of ECG signals.②It is researched to detect the ECG characteristic by down sampling. First, the method down sampling the ECG signals which has been filtered to reduce the data size of ECG signals. Second, detect P Wave, Q Wave, R Wave, S Wave, T Wave of every throb by amplitude and first difference threshold in down sampling ECG data. Finally, update every characteristic point by down sampling to finish detecting ECG characteristic. down sampling algorithm has no affect on the change of sampling frequency. Amplitude and first difference threshold use self adapting selecting technique. Second detection is processed by back tracking technique in potential omitting R wave. The techniques which mentioned above are able to obtain high percentage of accuracy and make a good foundation to followed analyzing.③It is researched to detect arrhythmia and wave classify. The method judge the throb normal or not through detecting difference time and difference amplitude in every characteristic of the throb, and the period between two adjacent throb. Classifying similar wave from every arrhythmia wave. Then hundreds of throbs are divided into tens of waves which have different characteristic each other. The method select threshold through patient's ECG wave characteristic and approach satisfied effect.④It is researched to build embedded platform. Embedded system has advantages such as running many hardware platform, clipping core, good performance, rich in software resource, low use-cost, strong network function , supporting graphics user interface and rich in development resource. The position is more and more obvious in developing medical software. Embedded platform use linux operator system. linux operator system opens its source code and owns heavy supporting population. The develop document is complete. It provides good develop platform for writing GUI program.⑤The Embedded ECG automatic analyzing software is developed. The software use qt to develop. It contains case history database manager, ECG wave display, ECG signals filter, ECG characteristic detecting, arrhythmia analyzing, wave classify, case report printing and so on. The software could deal ECG data which has difference sampling frequency or one to twelve channels. It also zoom out local ECG signals to view and edit for doctor. The algorithm which mentioned above is certificated by the software.