Optimal enhancement of the fetal electrocardiogram during labour

An important objective of this work is to make it possible to measure features from the fetal ECG on a beat-by-beat basis and so retain as much clinical information as possible. This has not previously been possible because quality of the raw fetal ECG data is often very poor. Baseline shift is the major cause of poor quality data but because it is usually caused by maternal movement, it is also only short lasting. Important features such as the ST segment shape can only be measured from good quality data and with minimal signal processing as attempts to enhance poor data can distort the ST shape. Other features, such as the RR and PR intervals are more robust and can tolerate higher degrees of signal processing. The authors feel it is very important that the quality of the raw data, as well as the features themselves, are measured on a beat-by-beat basis. This is made possible by use of curve fitting techniques which provide specific signal quality measures, as well as enhance the signal with minimal distortion. The authors are developing as part of a fuzzy expert system rules to decide what features are reliable based on these signal quality measures. Using this approach they believe their system can be robust enough to cope with the dynamic nature of labour and prevent the clinician from being mislead with false measurements