A new artificial intelligence model was developed to predict the age and gender of infants. The innovation works through an unusual kind of measurement, as it estimates the data based on the temperament of the kids.
Machine Learning System Predicts Infant Age and Gender
The machine learning algorithm was tested and proven to be effective at identifying the ages of infants of both genders respectively. According to the authors, the analysis successfully covered over 4,400 newborn babies.
The new method of scaling the ages and genders of infants through temperament is a novel solution that is quite challenging compared to modern measurements available today. The new AI-powered system, however, was able to differentiate the genders and even the ages of thousands of babies based on their temperament records documented in the first 48 weeks following their birth.
The tests during the first 48 weeks of infant babies allowed the AI model to learn the classification of the subjects based on the temperaments they exhibit. Through this improvement, experts concluded that the identification of gender differences and age in newborn individuals are more apparent during the particular window.
Washington State University's psychology specialist and lead author of the study Maria Gartstein explained in a TechXplore report that the approach they developed is the least suggestive but most effective way of estimating a child's age and gender.
Past investigations on temperament-based age and gender differences in babies showed promising solutions but were not able to consolidate both factors for prediction,
Gerstein said that the limiting steps in previous studies might have been due to the challenging phase of extraction of data from babies and the lack of resources in the few labs that carry out the research. This hinders the experts from getting ahold of statistically reliable and relevant information that could survey a broad population.
Hidden Importance of Temperament Data
The new AI-based research included datasets collected from babies, including behavior questionnaires, between the span of 2006 to 2019, IndiaTimes reported. The information was shared by supporting experts from various states.
The questionnaire considered in the study is a type of report filled out by parents. The measurement includes a spectrum of over 190 distinct behaviors that could be displayed by their offspring over three to 12 months of age.
The data is funneled into categories that could give one of 14 possible temperament dimensions that include emotional and physical responses such as smiling, agitation, frustration, and fear.
University of Idaho's Institute for Modeling, Collaboration, and Innovation expert Erich Seamon, who co-authored the study, managed the utility of the machine learning algorithms for the classification of infants who belong in the separate age groups of 0-24 weeks, 24-48 weeks, and 48 and over, based on the 14 temperament dimensions. The accuracy of data from the infants increased as their age went up, showing a striking improvement in the performance of the AI model's algorithms.
Among the findings, the authors say that fear is the prevalent key to categorizing male and female kids. Moreover, the effects of socialization on the children during this infancy period would be displayed after the kids turn a year old.
The study was published in the journal PLOS ONE, titled "Using machine learning to understand age and gender classification based on infant temperament."
RELATED ARTICLE: Artificial Intelligence Made Faster, More Efficient: New NVIDIA Processor Chip for AI Helps It to Better Understand Human
Check out more news and information on Artificial Intelligence in Science Times.
![Earth's Quasi-Moon Kamo‘oalewa Could Originate From Lunar Surface Not Asteroid Belt [Study]](https://1721181113.rsc.cdn77.org/data/thumbs/full/53275/89/56/50/40/earths-quasi-moon-kamo-oalewa-could-originate-from-lunar-surface-not-asteroid-belt-study.png)
