Posted in | Consumer Robotics

New AI Method can Identify and Distinguish Between Normal and Abnormal Cries of Babies

All parents are familiar with the annoyance of responding to a baby’s cries, thinking whether the baby is wet, hungry, tired, in need of a hug, or possibly even in pain.

A team of scientists from the United States has developed a new artificial intelligence technique that can recognize and differentiate between normal cry signals and abnormal ones, for example, those arising from an underlying illness. The technique, based on a cry language recognition algorithm, assures to be helpful to parents at home as well as in healthcare settings, since doctors may employ it to differentiate cries among sick children.

The study was reported in the May issue of IEEE/CAA Journal of Automatica Sinica (JAS), a joint publication of the IEEE and the Chinese Association of Automation.

Skilled health care workers and experienced parents can quite precisely differentiate between a baby’s several requirements depending on the crying sounds it creates. Although each baby’s cry is distinctive, they have some features in common when they stem from the same reasons. It has always been a major challenge to identify the hidden patterns in the cry signal, and artificial intelligence applications have now been established to be a suitable solution within this context.

The new study employs a particular algorithm on the basis of automatic speech recognition to identify and recognize the features of infant cries. To analyze and categorize those signals, the group employed compressed sensing as a way to process big data more efficiently. Compressed sensing is a process that reconstructs a signal based on meager data and is particularly useful when sounds are recorded in noisy environments, which is where baby cries normally occur.

In this research, the scientists developed a new cry language recognition algorithm that can differentiate the meanings of both normal and abnormal cry signals in a noisy environment. The algorithm does not depend on the individual crier, indicating that it can be used in a wider sense in practical cases as a way to identify and differentiate various cry features and better understand why babies are crying and how urgent the cries are.

Like a special language, there are lots of health-related information in various cry sounds. The differences between sound signals actually carry the information. These differences are represented by different features of the cry signals. To recognize and leverage the information, we have to extract the features and then obtain the information in it.

Lichuan Liu, Study Corresponding Author and Associate Professor, Electrical Engineering, CAA

Liu is the director of digital signal processing laboratory whose team carried out the study.

The scientists believe that the findings of their research can be applied to many other medical care circumstances in which decision making greatly depends on experience. “The ultimate goals are healthier babies and less pressure on parents and care givers,” stated Liu. “We are looking into collaborations with hospitals and medical research centers, to obtain more data and requirement scenario input, and hopefully we could have some products for clinical practice,” she added.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback