Doppler Audio Signal Analysis as an Additional Tool in Evaluation of Umbilical Artery Circulation

Research output: Contribution to journalArticle


Purpose: To investigate the predictive capacity of a new method for sound spectrum analysis of Doppler signals recorded from the umbilical artery in high-risk pregnancies. Material and Methods: The retrospective study comprised 127 pregnant women with various pregnancy complications between 23 and 39 gestational weeks. Umbilical artery blood flow velocity waveforms were recorded with Doppler ultrasound and characterized by pulsatility index (PI) and blood flow class (BFC). Doppler audio signals were stored on a digital video recorder and the sound frequency at the energy level 15 dB below its peak (MAXpeak-15 dB) was estimated off-line. The prediction of probability for composite adverse pregnancy outcome (operative delivery for fetal distress, admission to neonatal intensive care unit, perinatal death) was evaluated using the area under the curve (AUC) of the receiver operating characteristics (ROC) curve. Results: With increasing umbilical artery BFC, the MAXpeak-15 dB frequencies decreased (p < 0.0001) and the PI increased (p < 0.0001). The ROC AUCs for adverse outcome for MAXpeak-15 dB and for PI were 0.842 and 0.836 (p = 0.88), respectively. For the combination of MAXpeak-15 dB and PI, the corresponding AUC was 0.894, significantly higher than that of PI (p < 0.03) and of MAXpeak-15 dB (p < 0.05). Conclusion: Umbilical artery Doppler sound spectrum analysis might be a useful supplement to PI in the clinical evaluation of fetoplacental circulation.


External organisations
  • Skåne University Hospital
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Radiology, Nuclear Medicine and Medical Imaging
  • Obstetrics, Gynecology and Reproductive Medicine


  • blood velocity waveform, doppler audio signal, pregnancy, ultrasound Doppler, umbilical artery
Original languageEnglish
Pages (from-to)549-555
JournalUltraschall in der Medizin
Issue number5
Early online date2015 Sep 30
Publication statusPublished - 2017 Oct 1
Publication categoryResearch