Novel wearable ultrasound patch improves high-risk pregnancy monitoring

Current fetal monitoring tools for physicians to closely follow high-risk pregnancies are inadequate, so researchers at Stanford Medicine, the University of California San Diego and Oxford University developed a wearable ultrasound patch to monitor such pregnancies. The patch adheres to a patient's abdomen to provide continuous, real-time information about blood flow in the fetus and umbilical cord.

The device, whose design and early validation in several dozen pregnant patients was published May 26 in Nature Biotechnology, holds promise for helping doctors monitor conditions such as intrauterine growth restriction, which affects 10% of all pregnancies. In this pregnancy complication, the fetus grows slowly due to limited oxygen or nutrients, a result of insufficient blood flow through the umbilical cord. In severe cases, doctors deliver the baby early to enable better growth or avert stillbirth.

"There's nothing similar to our device on the market or in the scientific literature," said senior study author Sheng Xu, PhD, professor of anesthesiology, perioperative and pain medicine at Stanford Medicine. Xu was previously a professor at UC San Diego, where the majority of the research was conducted.

The study's lead authors are Geonho (Tom) Park, PhD, now a postdoctoral scholar in anesthesiology, perioperative and pain medicine at Stanford Medicine, who was a graduate student at UC San Diego when the research was conducted; as well as UC San Diego graduate students Yizhou Bian, Hao Huang and Sai Zhou.

Xu and Park are continuing to develop the ultrasound patch at Stanford Medicine.

The ultrasound patch is a flexible adhesive sticker, about the size of the palm of a hand, that adheres to the abdomen. It is connected by a cable to a computer that interprets ultrasound data. The researchers think the patch will initially be used for pregnant women who are hospitalized, but they hope eventually to have a wireless version that enables doctors to monitor patients at home.

"Umbilical artery blood flow is one of the components we look at closely when we're concerned about fetal well-being in cases of placental insufficiency," said Jane Chueh, MD, a high-risk obstetrician at Stanford Medicine who will be collaborating with Xu's team on further validation of the technology. But getting the data now is complex, she added.

The need for fetal data

Unlike the new ultrasound patch, current diagnostic tools usually show fetal status in small snapshots of time.

Not only does measuring blood flow on existing Doppler ultrasound machines capture data from short windows of time, but it requires a trained ultrasound technician and an appointment, Chueh said, complicating matters even for hospitalized patients.

Another current option for evaluating fetal well-being is cardiotocography, a combined measure of fetal heart rate and uterine contractions, data gained from a monitor strapped to the belly. Chueh's team deploys cardiotocography on women with high-risk pregnancies at Lucile Packard Children's Hospital Stanford.

"It's really hard to be on that continuously," Chueh said, noting that the equipment can give false signals or no signals when the fetus is moving a lot. "Even for inpatients, obtaining accurate readings three times a day can be labor intensive."

The monitor may show drops in fetal heart rate because of a real change, or just because the fetus moved around, requiring constant checks and repositioning of the monitor. "That's very, very stressful for both the patient and the medical staff," Chueh said.

More advanced monitoring devices would help physicians better walk the decision tightrope for when to deliver a baby in a risky pregnancy. Delivering earlier raises risks for complications of prematurity. But if the fetus isn't getting enough blood flow, waiting to deliver may be worse.

A technical challenge

In developing the patch, the researchers had to overcome a number of obstacles. Unlike most wearable devices, which measure information at or near the body's surface, this ultrasound patch needs to collect and interpret information from deep inside the uterus.

Also, everything they hoped to visualize is moving. Not only does the pregnant patient's body move, but the fetus practices its flip-turns, and the umbilical cord floats freely in the amniotic fluid. In a traditional Doppler ultrasound setup, a technician can reposition the machine's transducer to get a better view.

The researchers used a combination of strategy and technological innovation to tackle the problems.

We thought, 'What if we target the ultrasound device onto the placenta, in the area where the umbilical cord attaches?' Even though everything is moving, there is some stability in the umbilical cord at that location."

Geonho (Tom) Park, PhD, postdoctoral scholar in anesthesiology, perioperative and pain medicine, Stanford Medicine

The team developed an image-segmentation algorithm that can track the placenta-anchored end of the umbilical cord in real time, a key element of their design.

After developing a prototype and testing that it functioned as expected on a simulation mannequin, the researchers ensured it would not deliver too much acoustic or mechanical energy to a fetus. The device meets safety thresholds set by the U.S. Food and Drug Administration, the American Institute of Ultrasound in Medicine, and the British Medical Ultrasound Society.

The next step was to validate the device. The team tested the patch on 62 pregnant women, comparing its findings with those from a standard Doppler ultrasound machine. The new patch and the traditional machines produced statistically equivalent results.

The patch is able to image all three of the major blood vessels in the umbilical cord - two arteries and a vein. It also measures blood flow through a major artery in the fetus, and it can measure fetal anatomical structures such as head circumference, abdominal circumference and femur length to help estimate fetal weight, a key metric for diagnosing problems with growth. The image-segmentation algorithm correctly tracked the umbilical cord without repositioning on the mother's abdomen, even if mothers moved around. It also worked regardless of the location of the placenta within the uterus.

Detecting a serious complication

During the device validation, the research team saw something unexpected in one research participant.

"She was 28 weeks along in her pregnancy - still pretty early on - and an initial examination showed a normal fetal heart rate. Then I saw that the flow signal was quite abnormal," Park said. "I thought, 'Perhaps there's something malfunctioning in the device,' so I checked everything, but it seemed like the device was fine. I showed the data to the physicians who were there, and they agreed that the fetus might be in jeopardy."

The participant's ultrasound data showed large fluctuations in blood flow through the umbilical cord, in contrast to stable blood flow in a healthy patient at a similar stage of pregnancy.

Follow-up testing confirmed that the participant had a severe placental dysfunction. She was monitored closely by her physicians, and the baby was delivered by C-section four days later. The baby went to the neonatal intensive care unit and did well.

Next steps

The research team plans to further refine and validate use of the ultrasound patch at Stanford Medicine. In addition to building a wireless version of the device, they plan to test it in a wider range of patients with other pregnancy complications that involve poor blood flow, such as congenital heart disease and chronic hypertension.

"Right now, for these high-risk pregnant patients, it can be hard for physicians to get the information we want, right when we need it," Chueh said. "I think this device will be able to give us that information much more easily."

"We'll start with the inpatient population, but maybe someday we will also be able to use it on an outpatient basis," she added. 

The research was funded by grants from the National Institutes of Health (grants 1R01EB033464-01 and 1R01HL171652-01), as well as grants from Wellcome Leap and Accelerating Innovation to Market at UC San Diego.