Estimation of Fetal Blood Oxygen Saturation from Transabdominally Acquired Photoplethysmogram Waveforms.

Estimation of Fetal Blood Oxygen Saturation from Transabdominally Acquired Photoplethysmogram Waveforms.
复制标题

从经腹部获得的光电体积描记图波形估计胎儿血氧饱和度。

DOI:
10.1109/embc46164.2021.9629515
复制
发表时间:
2021
期刊:
Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
影响因子:
--
通讯作者:
Ghiasi,Soheil
Ghiasi,Soheil
中科院分区:
--
文献类型:
--
作者:
Vali,Kourosh;Kasap,Begum;Qian,Weitai;Vafi,Ata;Saffarpour,Mahya;Ghiasi,Soheil

文献摘要

相似文献

经腹胎儿脉搏血氧仪(TFO)面临着几个挑战,包括采集噪声光电容积描记图(PPG)信号,其中包含母体和弱胎儿信息的混合物,以及缺乏可以校准估计模型的数据点。本文提出了一种新的算法,解决了这些问题,并有助于胎儿血氧饱和度的估计,从PPG信号感测通过产妇腹部在一个非侵入性的方式。我们的方法由两个关键步骤组成。首先,我们开发的方法来近似的脉动和非脉动胎儿组织从感测到的混合信号的贡献。此外,我们利用有关所观察系统的先验信息,例如胎儿SpO2估计的生理可接受性,以减轻测量噪声并推断额外的数据样本,从而改进推断的SpO2估计模型。我们已经验证了我们的方法在体内,使用妊娠绵羊模型与缺氧胎儿羔羊。与从血气分析获得的金标准SaO2相比,我们的胎儿SpO2估计算法产生的交叉验证平均绝对误差(MAE)为6.29%,相关系数r=0.82。
Transabdominal Fetal Pulse Oximetry (TFO) faces several challenges, including the acquisition of noisy Photoplethysmogram (PPG) signals that contain a mixture of maternal and weak fetal information and scarcity of the data points on which an estimation model can be calibrated. This paper presents a novel algorithm that addresses these problems and contributes to the estimation of fetal blood oxygen saturation from PPG signals sensed through the maternal abdomen in a non-invasive manner. Our approach is composed of two critical steps. First, we develop methods to approximate the contribution of pulsating and non-pulsating fetal tissue from the sensed mixed signal. Furthermore, we leverage prior information about the system under observation, such as the physiological plausibility of fetal SpO2 estimates, to mitigate measurement noise and infer additional data samples, enabling improvements in the inferred SpO2 estimation model. We have validated our approach in-vivo, using a pregnant sheep model with a hypoxic fetal lamb. Compared with gold standard SaO2 obtained from blood gas analysis, our fetal SpO2 estimation algorithm yields the cross-validation mean absolute error (MAE) of 6.29% and correlation factor of r=0.82.