Interferometric near-infrared spectroscopy for transabdominal fetal oximetry
The outcomes and cost of childbirth is an important issue that almost every family faced or will face. Compared
to vaginal delivery, Cesarean section (C-section) has a higher cost, and may increase the health risk to both the
baby and mother. About 32% of the deliveries are via C-section, which is much higher than the 10-15% ideal
rate published by World Health Organization. A major triggering factor of C-section is a conservative prediction
of fetal hypoxia during labor, which unfortunately has a high false-positive rate using the existing, widely-adopted
intrapartum fetal monitoring technique of cardiotocography (CTG). Strikingly, since the introduction of CTG in
the early 1970s, the rate of C-section deliveries in the US has risen five folds, while the rates of conditions
associated with hypoxia remain unchanged. A noninvasive measurement of the fetal oxygen saturation during
labor and delivery could provide obstetricians specific indicator of the necessity of emergency C-section. This
may help to reduce the rate of unnecessary C-sections.
Here, we propose a new transabdominal fetal oximetry (TFO) technique based on an innovative interferometric
near-infrared spectroscopy (iNIRS) approach that could noninvasively measure the fetal arterial blood oxygen
saturation. The iNIRS TFO measures the time-resolved reflectance of near-infrared light shining on the maternal
abdomen. Our new method is distinctly different from the conventional pulse oximetry, and offers critical
advantage to detect weak fetal signals buried in strong overwhelming maternal signal through the abdomen while
greatly reducing the required light power on the tissue. Our approach distinguishes photons traversing different
tissue depth, thus facilitating the separation of signal between the shallow maternal layer and the deep fetal layer.
It greatly increases sensitivity to deep tissue signals and assures both an accurate and safe measurement.
We conducted preliminary experiment on pregnant ewe and verified our technique could detect the fetal
heartbeat through a single optical wavelength and a single detector, which presages the feasibility of iNIRS to
measure fetal oxygen saturation. Here, we aim to build a two-wavelength iNIRS with a significantly improved
measurement sensitivity to perform transabdominal fetal oximetry. The two optical wavelengths could extract the
relative change of the oxygenated and deoxygenated hemoglobin during a cardiac cycle, and we will use time-
division-multiplexing to switch between two optical wavelengths in the iNIRS setup. To improve the tissue
penetration depth, we will increase the frequency tuning rate of the laser, use an innovative detection fiber and
increase the number of detectors. All these approaches could increase the measurement sensitivity and thus
the accessible tissue depth. Finally, we will develop advanced machine learning algorithms to extract the fetal
oxygen saturation level from the raw measurements. We will perform the experiment in pregnant ewe model and
validate the results with the arterial blood gas (ABG) test. The success of our project will make a significant
impact on both the outcomes and cost of childbirth.