A routine heart test that millions of children already receive could quietly reveal far more than anyone realized: their biological maturation. Researchers at Wake Forest University School of Medicine in Winston-Salem, North Carolina have developed an AI-based tool that reads the electrical fingerprints of childhood development encoded in standard electrocardiograms—the same quick, non-invasive heart tests doctors order every day. The breakthrough challenges how pediatric researchers have long measured growth, replacing crude binary categories with a continuous spectrum that reflects how children actually develop.
The challenge pediatric researchers face is deceptively simple but vexing. Large medical studies often lack reliable information about children's pubertal stage or hormone levels, forcing researchers to rely on broad sex-based categories that flatten the gradual, uneven reality of biological development. Some children mature early; others late. Some show physical changes without corresponding hormone shifts, and vice versa. This gap in measurement has meant researchers studying cardiovascular health, drug responses, or developmental outcomes in huge populations must work with incomplete information. Until now.
The new tool, called the Electrocardiographic Sex Index or ESI, transforms routine ECG data into an AI-derived score that tracks maturation as a continuous process rather than fixed stages. When researchers applied an adult-trained ESI model to nearly 62,000 ECGs from children ages 0 to 18 drawn from the clinical archive at the University of Tennessee Health Science Center, a clear pattern emerged. In early childhood, ESI values clustered tightly together—boys and girls showed virtually no difference. But beginning in late childhood and accelerating through adolescence, the values began to diverge in opposite directions, reflecting the known biological differences that emerge during puberty. By mid-to-late adolescence, the pattern had stabilized, approaching adult levels.
What made the findings particularly robust was their consistency. The same age-related trends appeared across all racial groups studied, and the model's accuracy improved steadily as children aged, suggesting ESI captured something real about developmental biology rather than artifacts of the algorithm. "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," said Tolga Hayit, Ph.D., the study's co-lead author. "ESI offers a continuous measure that may help researchers account for developmental stage when Tanner staging or hormone data are not available."
The implications extend far beyond academic measurement. Large epidemiological studies involving millions of children currently treat sex as a binary category, missing the granular reality of how bodies change over time. For researchers designing studies of cardiovascular development, medication response, or long-term health outcomes, ESI could provide precision where only crude approximations existed before. The tool works on ECGs already collected in routine care—no new tests needed, no additional burden on families.
The study, published in the European Heart Journal—Digital Health, stops short of examining clinical outcomes. But it establishes what researchers call a foundation for future work examining how developmental maturity influences cardiovascular risk and treatment response over time. Next comes the deeper work: longitudinal studies that pair ESI scores with traditional pubertal staging and hormone measurements, validating whether this new continuous measure truly captures the complexity of childhood development. For now, researchers have revealed that the heartbeat itself whispers the story of growing up.