Implementation Research Of The ICA-based Fetal ECG Detecting And Monitoring System

Fetal Electrocardiogram (FECG) contains much physiological information. It can intuitively reflect the growth and health condition of perinatal fetus, so that the fetus’s diseases could be discovered and the treatment could be made in time. At present, there are two main methods of FECG detection, one is Fetal Scalp Electrode and the other is Maternal Abdominal Electrode. Fetal Scalp Electrode is an invasive detection method which can only be used during the production of the pregnant women, however, it will cause harm to the fetus and pregnant women. While Maternal Abdominal Electrode is a noninvasive detection method which can be easily and conveniently operated, what’s more, it won’t cause any harm to fetus and pregnant women. As a result, now it is a hotspot in the FECG monitoring area. Because the FECG is submerged in the large background noise, it can’t be used in fetal monitoring. Therefore, it’s very important to explore an effective method which can be used to extract clean FECG from mixture-signal.Independent Component Analysis (ICA) is a blind source separation algorithm which is based on the multidimensional signal statistical characteristics. The algorithm is to recover original source signals without any prior knowledge about the sources and mixing system. In recent years, ICA attracts widespread attention in the field of biomedical engineering. Many experts and scholars apply ICA to FECG extraction and they have obtained preferably separation results.Wireless sensor network is a wireless network which is consisted of many static or mobile sensors by the way of organization and multi-hop. It can sense the monitoring related information in the coverage of the area, and can also provide feedbacks. In view of the above features, it has a high application value in the health care field. Furthermore, home care which has significant scientific meaning and social worth will be an important trend of medical care development.This thesis is mainly about FECG monitoring system, particularly describes the design and implementation of the hardware and software module. The main research work and achievements of this thesis are as follows:1. The Infomax-algorithm is researched and applied to FECG extraction. Comparision of batch and online processing mode of Infomax-algorithm, and combination with ECG characteristics, consequently, the system use an online processing strategy in FECG extraction and obtain preferably experimental results. We put forward a new method of cumulant estimation based on sliding window and apply it in the implementation of online Infomax-algorithm. Meanwhile, the performance of online algorithm is improved dramatically.2. It has designed and implemented the ECG acquisition module. Major function of this module is signal amplifier and filter-pretreatment. It has solved the problem of FECG collection and preliminarily removed some noise interference, which can provide reasonable observation data to software module and conduct the follow-up analysis processing.3. Considering the information transmission and the different applications, both wired and wireless network communication module is designed and implemented.4. Subject to the laboratory conditions, simulation experiment is designed to collect FECG. Using the simulation-data and clinical-data verify the functions of each module of the system. The overall architecture of this system provides a feasible plan for implementation of FECG monitoring.