COMPUTING TECHNIQUES FOR INTRAPARTUM PHYSIOLOGICAL DATA REDUCTION .2. FETAL HEART-RATE

This paper describes a Computing technique which allows large quantities of Fetal Heart Rate (FHR) data to be reduced to a concise summary of the various components of FHR Variation. This summary, termed a RATEGRAM, was designed for clinical use in conjunction with the Powergram, s shown in Fig. 7. The first part of the processing is calculation of beat to beat intervals (measured in milliseconds) from the fetal EGG Signal by a program known as the SOFTWARE RATEMETER. The Software Ratemeter detects the Rwave peak of the ECG complex by adjusting an amplitude threshold in response to changes in amplitude and signal to noise ratio (S/N) of the signal. In addition, the authenticity of each beat to beat interval is validated, artefactual ones rejected and data quality assessed so that subsequent processing is not performed on poor quality data. Data reduction is based on generation of the RATE HISTOGRAM (RH) which contains the distribution of FHR during a ten minute epoch. Examples of Rate Histograms in Figs. l to 4 demonstrate that the RH accurately reflects the components of FHR activity. By plotting several descriptive parameters of the RH, epoch by epoch, a summary known as the RATEGRAM can be displayed, as shown in Figs. 5 and 6, which allows such components as: Baseline FHR level, Acceleration Activity, Deceleration Activity and Baseline Variability to be assessed. Baseline FHR level is plotted äs an ‘X’ in v those cases where it is certain, as a T where it is uncertain ‘and is not plotted when there is no baseline. The amount of acceleration and deceleration activity is indicated by the number of dots (.) above and below the baseline level. Occurrence of reduced baselinef variability is indicated by a X without accompanying dots. The performance of the Rategram and the likely data reduction were investigated using a ‘learning set’ of 7 labours and ‘test set9 of 18 labours consisting of a total of 1091 ten minute epochs, (approximately 182 hours) äs detailed in Tabs. I, II, III and IV. The majority of these labours were induced and/or augmented using oxytocic drugs. An ‘interesting’ epoch, that is, one which required further clinical action or study, was defined as one in which one or more of the following occurred: moderate or severe deceleration activity, reduced baseline variability, baseline bradycardia or tachycardia. Also included were those epochs where data quality was poor or a baseline could not be accurately defined. The number of interesting epochs within individual labours are detailed in Figs. l and 3, overall 25 per cent of epochs were clinically interesting and required further study or action. In conclusion, it is suggested that the use of data reduction and presentation techniques, such äs the Rategram, can highlight areas of clinically interesting FHR activity and can achieve a four-fold reduction in the amount of pen tracing requiring close clinical appraisal. © 1979, Walter de Gruyter. All rights reserved.