July 12, 2013
Cry Analyzer Detects Potential Health Problems In Babies
redOrbit Staff & Wire Reports - Your Universe Online
A new tool that analyzes slight variations in a baby's cry can help identify clues to potential health or developmental problems, giving doctors a new way to identify babies with neurological and other disorders, thus allowing for earlier intervention.
"There are lots of conditions that might manifest in differences in cry acoustics," said Stephen Sheinkopf, assistant professor of psychiatry and human behavior at Brown, who helped develop the new tool.
"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."
The cry analyzer, the result of a two-year collaboration between faculty at Brown's School of Engineering and Women & Infants Hospital, operates in two phases. During the first phase, the tool separates recorded cries into 12.5-millisecond frames. Each frame is analyzed for multiple variables, such as frequency characteristics, voicing and acoustic volume. The second phase uses data from the first to provide a more comprehensive view of the cry, and reduces the number of parameters to those that are most relevant.
The frames are then re-assembled and characterized as either an utterance or a silence (the pause between utterances). Longer utterances are separated from shorter ones, and the time in between recorded. Pitch, including the contour of pitch over time, and other variables can then be averaged across each utterance. In total, the system evaluates for 80 different parameters, each of which could hold clues about a baby's health.
"It's a comprehensive tool for getting as much important stuff out of a baby cry that we can," said Harvey Silverman, professor of engineering and director of Brown's Laboratory for Engineering Man/Machine Systems.
To determine which characteristics were most important in analyzing a baby's cry, Silverman and his team collaborated with Sheinkopf and Barry Lester, director of Brown's Center for the Study of Children at Risk.
"We looked at them as the experts about the kinds of signals we might want to get," Silverman said.
"We engineers were the experts on what we might actually be able to implement and methods to do so. So working together worked quite well."
Lester has spent years studying the cries of babies. He said this type of research dates back to the 1960s and a disorder known as Cri du chat (cry of the cat) syndrome, which is caused by a genetic anomaly similar to Downs syndrome. Babies who have the condition have a distinct, high-pitched cry.
While the Cri du chat is unmistakable to the human ear, Lester and others wondered whether subtler differences in cry could also be indicators of a child's health.
"The idea is that cry can be a window into the brain," Lester said.
If neurological deficits alter the way babies are able to control their vocal chords, those differences might also manifest themselves in variations in pitch and other acoustic features.
Lester has published several papers showing that differences in cry are linked to medical problems due to malnutrition, prenatal drug exposure and other risks.
"Cry is an early warning sign that can be used in the context of looking at the whole baby," Lester said.
The tools used in Lester's previous studies were primitive by today's standards, he said. Back then, recorded cries were converted to spectrograms, visual readouts of pitch changes over time, which were read by technicians and coded by hand. Later systems automated the process somewhat, although the research was still slow and burdensome.
The new automated cry analyzer allows researchers to evaluate cries much more quickly, and in far greater detail.
The Brown team plans to make the tool available to researchers around the world in the hopes of developing new avenues of cry research.
Sheinkopf said he plans to use the tool to search for cry features that might correlate with autism.
"We've known for a long time that older individuals with autism produce sounds or vocalizations that are unusual or atypical," said Sheinkopf, who specializes in developmental disorders.
"So vocalizations in babies have been discussed as being useful in developing early identification tools for autism. That's been a major challenge. How do you find signs of autism in infancy?"
The answer may very well be encoded in a cry.
"Early detection of developmental disorders is critical," Lester added. "It can lead to insights into the causes of these disorders and interventions to prevent or reduce the severity of impairment."
A paper describing the cry analyzer is published in the Journal of Speech, Language and Hearing Research.