AI model reveals hidden sign of puberty in routine heart tests, tracking child maturity patterns

AI model tracks child maturity patterns through routine ECG data

Wake Forest University School of MedicineMay 28 2026 A new study from Wake Forest University School of Medicine suggests a routine heart test - an electrocardiogram (ECG) - may offer researchers a new way to measure biological development in children and adolescents. The findings were published in a study in European Heart Journal - Digital Health. Pediatric researchers often lack reliable measures of pubertal stage or hormone levels in large datasets, forcing them to rely on broad sex-based categories that may not fully capture the gradual nature of biological development. To address this, researchers developed the Electrocardiographic Sex Index (ESI), an AI-based score derived from standard ECGs that reflect biological development on a spectrum rather than in fixed categories. In a study of more than 60,000 children's ECGs, researchers found the score changed in predictable way as kids grow. In early childhood, the values were very similar among children. But as they reach later childhood and the teen years, the values begin to separate in ways that reflect normal growth and hormonal changes. These findings suggest that ESI captures the step-by-step changes of normal development, rather than a simple split into categories. This approach could give researchers a more precise way to account for developmental stage especially when hormone or puberty data isn't available. One of the most exciting aspects of this work is it shows routine ECG data may contain meaningful information about biological maturation in children and adolescents. ESI offers a continuous measure that may help researchers account for developmental stage when Tanner staging, a standard medical system used to describe the five stages of puberty, or hormone data, are not available." Tolga Hayit, Ph.D., visiting researcher and study's co-lead author "ECGs, traditionally underutilized for capturing developmental biology, can now, when coupled with state-of-the-art AI approaches, highlight their potential to uncover patterns of maturation and cardiovascular development at scale," said Ibrahim Karabayir, Ph.D., assistant professor of cardiology and at the Wake Forest Center for Artificial Intelligence Research and the study's other co-first author. Key findings * In early childhood, ESI values were tightly centered, showing little difference between children grouped by biological sex. * Beginning in late childhood and becoming more pronounced through adolescence, ESI values diverged in opposite directions, plateauing in mid‑to‑late adolescence. * The same age-related trends were observed similarly among all races. * As the children got older, the model's accuracy steadily improved with children approaching adult-level performance in older adolescents. To conduct the study, researchers applied the adult‑trained ESI model to 61,930 ECGs from children ages 0-18 years drawn from the clinical ECG archive at the University of Tennessee Health Science Center. The model was applied without retraining or recalibration, allowing investigators to directly observe how ECG features evolve relative to adult benchmarks. Why it matters Many studies involving children treat sex as simple category, which doesn't always fully capture how their bodies develop overtime. ESI may offer a better way to account for these gradual, stage-by-stage changes as children grow, the study's authors said. For researchers working on large studies, ESI could be helpful when information about puberty or hormone levels aren't available. Next steps The researchers highlight the need for longitudinal studies incorporating Tanner staging, a standard five stage clinical scale used to access physical development during puberty hormone measurements and outcomes to further evaluate the clinical and biological significance of ESI in pediatric populations. While the study does not assess clinical outcomes in children, it establishes a foundation for future research examining how developmental maturity influences cardiovascular risk, treatment response or long‑term outcomes - using ECGs already collected in routine care. Future studies should evaluate ESI longitudinally and incorporate Tanner staging, hormone measurements and clinical outcomes to better understand its biological and clinical significance in pediatric populations. Source: Wake Forest University School of Medicine Journal reference: Hayit, T., et al. (2026) ECG Sex Index in children and adolescents. European Heart Journal - Digital Health. DOI: 10.1093/ehjdh/ztag058. https://academic.oup.com/ehjdh/article/7/4/ztag058/8651698

Scientists found a hidden sign of puberty in routine heart tests

Pediatric researchers face a familiar problem. Most large studies treat children as one of two groups - boys and girls - and leave the gradual reality of development unseen. A standard heart recording may offer something better. An electrocardiogram - the same test ordered in any routine checkup - carries a hidden signal that tracks how a child matures, without the need for a blood draw. Electrical patterns in the heart Researchers at Wake Forest University School of Medicine developed a way to read that signal from an ordinary electrocardiogram, using artificial intelligence to analyze the heart's electrical patterns. The approach condenses an entire recording into a single score. Scores run from 0 to 1 on what the team calls the Electrocardiographic Sex Index, or ESI. Readings near 1 carry electrical features typical of boys, while those near 0 match girls. The idea began with adults, where the index sorts male from female hearts with ease. Dr. Tolga Hayit, co-lead author of the study, wanted to see what it would do with children's hearts. A test on children The team borrowed a huge archive of children's heart tests from the University of Tennessee Health Science Center (UTHSC) in Memphis. The collection held nearly 62,000 ECGs from kids between birth and age 18. Rather than build a fresh model for young patients, the researchers ran the adult version on every child's recording untouched - no retraining and no adjusting the dials for smaller, faster hearts. That choice let them watch one thing closely: the age when a child's heartbeat starts to read like an adult's heartbeat. If the model struggled with toddlers but performed well in teenagers, that alone would reveal when the heart begins taking on adult-like electrical patterns. A line that splits In the youngest children, the scores barely budged. Boys and girls alike clustered right around the middle of the scale, near 0.5. Their heartbeats were nearly impossible to tell apart on electrical grounds. Then came the teen years. The two groups pulled apart in opposite directions - boys' scores climbing toward 1, girls' drifting toward 0. The widening gap held steady into the late teens. This is the part no one had charted before. Built for adults, the index still followed the step-by-step divergence that puberty brings - doing so in Black and White children alike, with only small differences between them. Sharper differences every year The model's accuracy told the same story. With newborns it could barely tell boys from girls, yet by age 17 it was right almost every time. That steady climb mirrors what doctors already see in the body. Electrical timing differences between boys and girls stay faint through childhood. Then hormones rise - and the gap widens, a pattern documented in an earlier review. One known example shows up right on the paper. After puberty, girls tend to have a slightly longer pause between heartbeats than boys - a difference barely there at birth that only appears once the body matures. An electrical signal of puberty A skeptic might guess the model was just tracking heart rate, since kids' hearts slow with age. The team checked and found almost no link at all. On its own, heart rate could not tell boys from girls - no better than flipping a coin. The index could, which means it was reading something deeper in the signal than simple speed. What the index captures, though, is not puberty itself. It reads the electrical pattern of the heartbeat. The team can only say that pattern changes with age in step with growing up, not that it tracks hormones. Potential medical applications Pediatric researchers hit a stubborn wall. They rarely know where each kid stands in development, so they lean on a blunt boy-or-girl split that misses the gradual truth. Hormones slide along a smooth curve from infancy onward rather than jumping between fixed boxes, as age-based developmental charts confirm in one study. A single number that tracks that curve could give researchers a far finer tool. The clearest payoff may be in children treated for cancer. Chemotherapy and chest radiation can strain the heart for years, and a score marking which patients already have grown-up hearts could help doctors weigh that risk more precisely. Future research directions Until now, no one had shown that a heart test built for adults could follow childhood without being rebuilt. This work did just that, turning a routine scan into a record of how a young body matures. For Dr. Hayit, that is the surprise worth sitting with. "Routine ECG data may contain meaningful information about biological maturation in children and adolescents," said Dr. Hayit. What the study cannot yet answer is whether a child's score predicts future health. That will require following children over time and comparing their ECG scores with hormone levels, growth patterns, and later medical outcomes. The study is published in The European Heart Journal. -- - Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com. -- -

Routine Heart Test Can Track How Kids Grow and Mature, New Study Finds | Newswise

Approach uses AI-analysis to track changes in growth and puberty without relying on hormone data Newswise -- WINSTON-SALEM, N.C. --  A new study from Wake Forest University School of Medicine suggests a routine heart test - an electrocardiogram (ECG) - may offer researchers a new way to measure biological development in children and adolescents. The findings were published in a study in European Heart Journal - Digital Health. Pediatric researchers often lack reliable measures of pubertal stage or hormone levels in large datasets, forcing them to rely on broad sex-based categories that may not fully capture the gradual nature of biological development. To address this, researchers developed the Electrocardiographic Sex Index (ESI), an AI-based score derived from standard ECGs that reflect biological development on a spectrum rather than in fixed categories. In a study of more than 60,000 children's ECGs, researchers found the score changed in predictable way as kids grow. In early childhood, the values were very similar among children. But as they reach later childhood and the teen years, the values begin to separate in ways that reflect normal growth and hormonal changes. These findings suggest that ESI captures the step-by-step changes of normal development, rather than a simple split into categories. This approach could give researchers a more precise way to account for developmental stage especially when hormone or puberty data isn't available. "One of the most exciting aspects of this work is it shows routine ECG data may contain meaningful information about biological maturation in children and adolescents," Tolga Hayit, Ph.D., a visiting researcher and the study's co-lead author, said. "ESI offers a continuous measure that may help researchers account for developmental stage when Tanner staging, a standard medical system used to describe the five stages of puberty, or hormone data, are not available." "ECGs, traditionally underutilized for capturing developmental biology, can now, when coupled with state-of-the-art AI approaches, highlight their potential to uncover patterns of maturation and cardiovascular development at scale," said Ibrahim Karabayir, Ph.D., assistant professor of cardiology and at the Wake Forest Center for Artificial Intelligence Research and the study's other co-first author. Key findings * In early childhood, ESI values were tightly centered, showing little difference between children grouped by biological sex. * Beginning in late childhood and becoming more pronounced through adolescence, ESI values diverged in opposite directions, plateauing in mid‑to‑late adolescence. * The same age-related trends were observed similarly among all races. * As the children got older, the model's accuracy steadily improved with children approaching adult-level performance in older adolescents. To conduct the study, researchers applied the adult‑trained ESI model to 61,930 ECGs from children ages 0-18 years drawn from the clinical ECG archive at the University of Tennessee Health Science Center. The model was applied without retraining or recalibration, allowing investigators to directly observe how ECG features evolve relative to adult benchmarks. Why it matters Many studies involving children treat sex as simple category, which doesn't always fully capture how their bodies develop overtime. ESI may offer a better way to account for these gradual, stage-by-stage changes as children grow, the study's authors said. For researchers working on large studies, ESI could be helpful when information about puberty or hormone levels aren't available. Next steps The researchers highlight the need for longitudinal studies incorporating Tanner staging, a standard five stage clinical scale used to access physical development during puberty hormone measurements and outcomes to further evaluate the clinical and biological significance of ESI in pediatric populations. While the study does not assess clinical outcomes in children, it establishes a foundation for future research examining how developmental maturity influences cardiovascular risk, treatment response or long‑term outcomes -- using ECGs already collected in routine care. Future studies should evaluate ESI longitudinally and incorporate Tanner staging, hormone measurements and clinical outcomes to better understand its biological and clinical significance in pediatric populations. About Wake Forest University School of Medicine Wake Forest University School of Medicine is the academic core of Charlotte, North Carolina-based Advocate Health and a recognized leader in experiential medical education and groundbreaking research. It directs the education of nearly 1,900 students and fellows, including physicians, basic scientists and allied clinical professionals. The school of medicine also strategically investigates opportunities that will expand basic and clinical research, resulting in nationally and internationally recognized excellence in biomedical research. The school has two campuses, each co-located with leading-edge innovation districts, The Pearl, in Charlotte, and Innovation Quarter, in Winston-Salem, North Carolina. These affiliated life-sciences innovation districts focus on advancing health care through new medical technologies and biomedical discovery. About Advocate Health Headquartered in Charlotte, North Carolina, Advocate Health is the third-largest nonprofit, integrated health system in the United States. A preeminent academic health system at the forefront of clinical excellence, innovation and research, it delivers care under the names Advocate Health Care in Illinois; Atrium Health in the Carolinas, Georgia and Alabama; and Aurora Health Care in Wisconsin and Michigan, and Wake Forest University School of Medicine is its academic core. Nationally recognized for expertise in heart and vascular, neurosciences, oncology, pediatrics and rehabilitation, Advocate Health is also a pioneer in the delivery of virtual health care. It is accelerating discovery by making research participation part of the standard-of-care through its one-of-a-kind National Center for Clinical Trials, plus two affiliated life-sciences-focused innovation districts and one of the nation's largest graduate medical education programs. With more than 165,000 teammates serving patients at 69 hospitals and over 1,000 care locations across eight states, Advocate Health reinvests over $6 billion each year to improve community health, making it one of the nation's largest providers of community benefit.