QNRF Newsletter Archive

Qatar-based research could revolutionize fetal and newborn healthcare

Professor Boualem Boashash of Qatar University
An expert in signal analysis, Professor Boualem Boashash, of Qatar University’s College of Engineering, worked for the US and Australian military for years. But over time, he decided he’d rather apply his skills to a different sector of society.

“In terms of my personal history, if you want to move from work that is with the military, it’s best to move to work that is for the health of infants, the most innocent, Dr. Boashash said. “So in 1992 I had a meeting with a colleague in Australia, and we looked at potential collaborations. We brainstormed, and since then we made a lot of progress.”

The research team—Dr. Boashash and colleagues from the University of Queensland’s UQ Centre for Clinical Research, Herston, Australia, and the University of Melbourne’s Department of Obstetrics and Gynaecology—are working to monitor the health of developing fetuses using signal detection. The signals, based on movements of a fetus inside the womb, can be read over time to detect any unusual patterns or a reduction in movement. But this requires a level of expertise based on signal-analysis and interpretation techniques that have been largely overlooked until now.

“There are two major variables in a signal,” Dr. Boashash explained, “time and frequency. The way our approach differs from classical techniques is that we are going to use these two variables concurrently. This is major, because in the past either time delineated or frequency information was used, but not the two combined. It would be like trying to represent a person by just looking at the height while the person’s weight and other factors are ignored.”

Because of this isolation of frequency from time, Dr. Boashash said, most signal processing techniques make an assumption that the spectral components in the signal do not change over time.

“In fact this assumption is just an approximation,” he said. “Without this assumption, we can employ more advanced techniques that allow us to extract more information. If we have more information we can make better decisions.”

Through this integrated approach to signal analysis, Dr. Boashash said he is able to look at signals represented as “three dimensional energy, or amplitude, as a function of time and frequency.” This is to say that the signals provide a more complete picture of the quality of the movements that produced them, and this allows the researchers to draw meaning from the signal changes.

The movement of an unborn child is a key indicator of its health, Dr. Boashash said. So the team has developed a sensor system that, when placed on the mother’s stomach, detects signals from movements inside the womb. The nature of these signals can be determined to detect any abnormalities in the fetus’ movement. So that the mother’s movement is not mistaken for that of the fetus, a separate sensor is positioned on her back and those signals are read separately.

“Mothers can feel about a third of the movements made by the unborn baby and can often detect a change in the movement patterns,” Dr. Boashash said. “However, because this varies from baby to baby, and woman to woman, some changes in movement patterns may go undetected. Sadly, some babies die in the womb following a period of time of decreased movements. Essentially, the aim of this project is to infer from recordings of the fetal movements what the health of the fetus is, and try to make a decision before it is too late.”

Dr. Boashash explained that the technology allows for continuous monitoring of the fetus so that any unusual signal patterns will be sent to the physician. This would eliminate the need for frequent visits to the doctor and would at the same time monitor the health of the baby more effectively.
Newborn wearing monitors

A newborn connected to frequency monitors
The team is also working on newborn monitor systems that work on the same premise—signals from the brainwaves of the infant are transferred into sensors placed in a “hat” worn just after birth.

“It has electrodes that measure the electrical activity at the scalp, and the electrical activity on the scalp is directly related to the activity of the neurons inside the brain,” Dr. Boashash explained. “The chemical reactions produce electricity, just like a battery produces electricity based on chemicals inside it.”

This technology again relies on a more sophisticated approach to the signals. The peaks and frequency of the signal offer some information, Dr. Boashash said, and the order of these peaks in time offers additional information.

“We have shown that when a newborn has a seizure, there is a particular pattern when you look at frequency and peaks together,” he said. “You will not see that there is a problem without that pattern. The novelty of our work is that when a medical doctor looks at the baby, he may recognize that the baby is going through a seizure pattern because of the results given the way the signals have been processed.”

The research is especially important in Qatar, Dr. Boashash said, where monitoring of fetuses and infants could prevent problems into the future, and where the privacy of such technologies would encourage more families to use them.

“Women here could benefit from the technology because they are less likely to want to go and show their tummy to a doctor,” Dr. Boashash said.

Dr. Boashash and his team have published their findings widely and are currently testing the monitoring systems in Australia with the intention of testing it in the near future in Qatar.

NPRP: 09 - 465 - 2 - 174
EEG signal classification for newborn outcome prediction after perinatal brain injury.
NPRP: 09 - 626 - 2 - 243
Design of a computer-automated system for foetal movement classification in the case of high-risk pregnancies using advanced signal processing algorithms.

Number of views (2099)

Please login or register to post comments.