Analysis of fetal movement based on magnetocardiographically determined fetal actograms and fetal heart rate accelerations

Abstract

Recently, we developed a method for the automatic detection of gross fetal movements using multichannel fetal magnetocardiogram (FMCG) signals. This approach exploits the changes in fetal QRS morphology in specific channels when the fetal heart changes its position relative to those channels. Aim of this work was to assess this approach in a group of subjects with respect to heart rate accelerations through the third trimester of pregnancy. 26 FMCGs were recorded for 5 min. in 14 healthy pregnant women (31st–41st week of gestation) using a 61 channel biomagnetometer. In each recording, after the removal of the maternal artifact, the fetal QRS complexes were identified using template matching. In each channel, the minimum and maximum amplitudes of each QRS complex were determined. In the actogram, the beat-to-beat changes in maximum and minimum values, respectively, were quantified using the variance over a 20 beat window. Automatically determined, rapid increases in variance were considered reflecting changes in fetal position. The timings of these increases were examined together with changes in fetal heart rate (determined by the RR intervals derived from the QRS complexes). We found between 1–4 episodes of rapid increases in variance in each of the 26 actograms. These episodes were associated with an immediate decrease of 31.7±18.8 ms from overall RR interval (heart rate increase: 11.3±6.5 bpm). Beat-to-beat RR interval variability (root mean square of successive differences) was reduced from overall 7.4±3.2 ms to 3.8±1.7 ms immediately following the begin of the actographically determined episodes. The results suggest that the majority of the episodes reflected fetal movement and that this actographic approach may be used to automatically scan FMCG recordings for such movement. This will be helpful in categorizing FMCG recordings with respect to fetal behavioral state and improve the assessment of fetal heart rate variability.