Voice Recordings May Help Identify Autism, Study Says
New technology that analyzes recordings of children's voices may be able to help diagnose autism, a recent study shows.
The Language Environment Analysis (LENA) system is attached to children's clothing and records sounds they make in a day. Researchers developed an automated system that can evaluate the vocalizations of very young children and identify a child with autism or a language delay, according to the study published in the Proceedings of the National Academy of Sciences.
The development of the technology was funded by the previous owners of Infoture, Inc., a for-profit company that was dissolved in 2009 and reconstituted as the nonprofit LENA Foundation. The study's authors had been on the Scientific Advisory Board of Infoture and had received consultation fees from the company before it folded.The researchers analyzed nearly 1,500 recordings made by 232 children between the ages of 1 and 4, who wore the devices all day while going about their regular activities. The recorder fits into the pocket of specially designed clothing.
The most important indication of vocal development was the ability of a child to produce well-formed syllables with rapid movement of the jaw and tongue. The analysis is based on sound patterns, not words.
The device and speech analysis make it possible to identify a child with an autism spectrum disorder much earlier than is currently the norm, the authors write. The technology could be used as a screening device by pediatricians, who could then refer children to a specialist for a full diagnosis.
Related: Researchers Blast Another Link Between Autism and Vaccines
Start by teaching him that it is safe to do so.