Alexander Kazantsev1, Julia Ponomareva2, Elena Chatskis3, Daria Medvedeva4, Nickolay Minaev4, Andrey Senin1, Pavel Kazantsev5, Lilya Subbotina4
09:05 - 09:55 | Fri 17 Feb | Ballroom D | FrRAF.22
The final goal of this work is an affordable large-scale technology for fetal monitoring to eradicate prenatal morbidity and mortality. Initially it is aimed at sonography automation for early fetal monitoring at the second trimester of pregnancy. Off-the-shelf fetal Doppler and smartphone-enabled implementation of a low cost in-home monitoring system are the keys to our approach. A grid of neural networks was designed for signal processing automation in a personal hemodynamics monitoring system. The results: 99% reliability of cardiac cycle detecting; 95% of fetal presentation recognition; 1.5% accuracy of diastolic ratio evaluation.