CNET-‘Cry analyzer’ helps identify neurological or developmental disorders in infants
‘Cry analyzer’ helps identify neurological or developmental disorders in infants
July 11, 2013 | Elizabeth Armstrong Moore
The computer-based tool out of Brown University performs finely tuned acoustic analyses across 80 parameters that reveal details about a baby’s health.
A new tool out of Brown analyzes the acoustic signal of a baby’s cry to alert caregivers of potential problems.
(Credit: Mike Cohea/Brown University)
Much as we parents like to think we know our babies best, subtle clues lurk in their cries that are, for the most part, imperceptible to the human ear, and they can reveal important information about a child’s health.
So researchers at Brown University and Women & Infants Hospital of Rhode Island hope their new computer-based cry analyzer will help researchers, clinicians, and caregivers identify possible neurological or developmental issues at a very young age.
“There are lots of conditions that might manifest in differences in cry acoustics,” developer Stephen Sheinkopf, assistant professor of psychiatry and human behavior at Brown, said in a school news release. “For instance, babies with birth trauma or brain injury as a result of complications in pregnancy or birth, or babies who are extremely premature, can have ongoing medical effects. Cry analysis can be a noninvasive way to get a measurement of these disruptions in the neurobiological and neurobehavioral systems in very young babies.”
Reporting in the Journal of Speech, Language, and Hearing Research, researchers say their tool operates in two phases. In the first, an analyzer cuts the recording of a baby crying into 12.5-millisecond chunks and analyzes them for a variety of parameters, including acoustic volume and how frequently certain characteristics occur. During the second phase, the frames are put back together and characterized using info from the first phase as a single “wah” or as silence — the pause between the wahs. The time between those utterances is recorded, and all of this allows for things like pitch and other variables to be averaged across utterances.
Ultimately, the tool is able to evaluate 80 parameters, providing an auditory view of the child’s brain. Barry Lester, director of Brown’s Center for the Study of Children at Risk, has studied baby cries for years and says the research actually dates back to the ’60s and the Cri du chat syndrome — similar to Downs and marked by a distinct, high-pitched catlike cry that can be detected even without specific tools. The finding prompted further investigation into what a baby’s cries can reveal about his or her health.
“The idea is that cry can be a window into the brain,” Lester said. Neurological disorders can actually affect how a baby controls his or her vocal chords. Lester has also published several papers revealing that a baby’s cry can be linked to issues that stem from a range of factors, including prenatal drug exposure and malnutrition.
‘Smart Diaper’ with QR code may track tots’ health
TweetPee: Huggies sends a tweet when baby’s wet
Wireless network could monitor breathing
The new analyzer gives researchers a tool to evaluate cries far faster and in greater detail than previous systems that included messy visual readouts of pitch changes over time that technicians had to code by hand.
The team at Brown plans to make its tool available to researchers around the world as they continue to establish links between cries and disorders, with autism being a particular area of interest.
“Early detection of developmental disorders is critical,” Lester said. “It can lead to insights into the causes of these disorders and interventions to prevent or reduce the severity of impairment.”
Best of all perhaps, the approach is completely noninvasive.
Elizabeth Armstrong Moore
Elizabeth Armstrong Moore is based in Portland, Ore., and has written for Wired, The Christian Science Monitor, and public radio. Her semi-obscure hobbies include climbing, billiards, board games that take up a lot of space, and piano.