For kids with speech and language disorders, early intervention can make a world of difference. But according to research at MIT and Massachusetts General Hospital, 60 percent of these children may not even be diagnosed until they’re in Kindergarten. Sometimes it takes even longer than that. At that point, those disorders can have a huge negative impact on academic and social success.

In order to facilitate early intervention for young children with speech or language disorders, the research team is developing software that can detect these issues, giving caregivers an opportunity to further investigate and treat them.

While it’s fairly obvious when children have a speech impediment due to a cleft palate or other physical issues, speech and language disorders are neurological in origin. Jordan Green, one of the researchers, explains that speech disorders affect motor nerve pathways while language disorders affect cognitive and linguistic pathways. Impaired motor nerve pathways would make it difficult for the tongue to actually make the shapes required for proper speech, while impaired cognitive and linguistic pathways would make it difficult for children’s brains to find the words they want to say.

The software works by giving children a storytelling test and listening for 13 acoustic signifiers that indicate such a disorder. It’s based on the idea that pauses in children’s speech would be a telltale indicator of a speech or language problem. It has been proven at this point to be accurate enough to be useful in detecting these issues.

“The really exciting idea here is to be able to do screening in a fully automated way using very simplistic tools,” says John Guttag, the senior author of the paper. “You could imagine the storytelling task being totally done with a tablet or a phone. I think this opens up the possibility of low-cost screening for large numbers of children.”

The software is still very much in the experimental phase, and has so far been built on existing data, but has proven quite effective at detecting the signifiers in question.

Thomas Campbell, a professor at the University of Texas at Dallas believes “the researchers’ automated approach to screening provides an exciting technological advancement that could prove to be a breakthrough in speech and language screening of thousand of young children across the United States.”