Ultrasound Bone Age Test Matches X-Rays Without Radiation in Children

For more than half a century, the standard way to measure how mature a child’s skeleton has become has been to take an X-ray of the left hand and wrist and compare it against a reference atlas of growing bones. The technique works, and it remains the gold standard in clinics around the world, but it carries a drawback that matters enormously in pediatrics: every assessment delivers a small dose of ionizing radiation to a growing body. Now a team of researchers in China has refined a radiation-free alternative that may rival the X-ray method with startling precision, using nothing more than an ultrasound probe, a simple ratio, and three bones of a child’s arm.

The study, published in Pediatric Radiology, comes from Wuhan Children’s Hospital and its affiliated medical college, where a prospective cross-sectional trial enrolled 144 consecutive outpatients between September 2023 and September 2024. The cohort was deliberately balanced, with 72 boys and 72 girls, and covered an extraordinarily wide developmental window: boys ranged in chronological age from 0.2 to 17.8 years, and girls from 0.4 to 17.8 years. Every child underwent both a conventional left-wrist radiograph and a multi-joint ultrasound examination of the wrist, elbow, shoulder, and knee, allowing the researchers to compare their new ultrasound parameter directly against the established radiographic benchmark.

The technical heart of the method lies in a quantity the researchers call the ossification ratio, or OR. On an ultrasound image, each ossification center appears as a discrete, variably sized echogenic focus wrapped in a circumferential halo of hypoechoic cartilage. The sonographer measures the maximum epiphyseal diameter, denoted D, and the diameter of the ossification center within it, denoted d, along the longitudinal axis of the bone. The ratio of these two values, d divided by D, expresses how far that particular growth center has progressed toward full ossification. Because cartilage and bone reflect sound waves differently, ultrasound can visualize these structures even before they are visible on X-ray, and it does so without any radiation exposure at all.

In the study, seven ossification centers and epiphyses were evaluated: the proximal first metacarpal, the distal radius, the distal ulna, the capitellum of the humerus, the proximal radius, the proximal humerus, and the medial epicondyle of the femur. All scans were acquired with a GE LOGIQ e system equipped with a 7.0 to 12.5 MHz high-frequency linear-array transducer, following standardized probe positioning protocols adapted from previous work. A single musculoskeletal sonographer with eight years of pediatric experience performed every scan while remaining blinded to each participant’s chronological age and radiographic bone age, a design feature intended to eliminate expectation bias from the measurements.

The headline result is a correlation so strong it borders on the theoretical limit. When the researchers summed the ossification ratios of just three bones, the distal radius, the distal ulna, and the proximal radius, the resulting composite parameter correlated with radiographic bone age, as assessed by the Greulich-Pyle atlas, at a Pearson coefficient of 0.98 in boys and 0.97 in girls. For context, a coefficient of 1.0 would indicate perfect correlation. Strikingly, adding the ossification ratios of other bones did not improve the correlation, and summing the ratios of all seven assessed bones actually reduced it slightly, to 0.97 in boys and 0.96 in girls. The three-bone combination appears to hit a sweet spot of information density without diluting the signal.

Reproducibility, the other pillar of any clinical measurement, proved equally impressive. The researchers evaluated both intra-observer and inter-observer reliability using intraclass correlation coefficients, with two blinded orthopedic surgeons of differing experience levels repeating measurements on images two months apart. For the summed three-bone parameter, intra-observer reliability reached an ICC of 0.99, with a 95 percent confidence interval of 0.98 to 0.99, and inter-observer reliability was also 0.99. Both values were significantly higher than those of any single sonographic parameter. The first metacarpal, by contrast, showed the weakest correlation with bone age of any individual bone, at 0.80 in boys and 0.77 in girls, and the lowest inter-observer reliability at an ICC of 0.84, underscoring why the composite approach outperforms single-bone measurements.

The choice of the proximal radius as a component of the index was not arbitrary. Among all seven bones examined, the proximal radius showed the strongest individual correlation with radiographic bone age, at 0.95 in boys and 0.94 in girls. This aligns with earlier forensic and orthopedic research showing that proximal radial epiphyseal fusion scores correlate most strongly with chronological age among elbow landmarks. Ossification of the proximal radius begins between three and six years of age, with the ossification center fusing at a mean age of 14.8 years in boys and 12.5 years in girls, making it a powerful developmental marker across a broad age range. The humeral capitellum and proximal humerus, by contrast, showed only moderate correlations of roughly 0.81 to 0.87 and were excluded from the final index.

The practical numbers behind the new parameter are easy to grasp. In adolescents with a bone age of 14 to 16 years, boys in the study exhibited ossification ratios of 82.3 percent for the distal radius, 61.7 percent for the distal ulna, and 79.1 percent for the proximal radius, summing to 232 percent. Girls in the same bone-age group showed corresponding values of 90.1, 81.6, and 76.1 percent, for a total of 247.5 percent. By ages 16 to 18, individual ratios approached 80 to 90 percent and the composite sum reached 250 to 260 percent, reflecting near-complete ossification as skeletal maturity arrives. These reference values, organized by bone-age group and sex, give clinicians a lookup framework for translating an ultrasound measurement into an estimated skeletal age.

Why does eliminating radiation matter so much? Although the effective dose from a single wrist radiograph is very low, children with growth disorders, short stature, or idiopathic scoliosis frequently require bone age assessments multiple times per year over many years, and the cumulative exposure adds up. Magnetic resonance imaging can assess skeletal maturity without ionizing radiation, but its high cost, long scan times, and limited accessibility make it impractical for routine repeated use. Ultrasound, by contrast, is inexpensive, widely available, well tolerated by children, and fast. The new three-bone index offers a further logistical advantage over a previously proposed composite that combined the radius, ulna, and femur: it requires only a single upper-limb scan, eliminating femoral imaging and repeated repositioning, which shortens preparation time, reduces discomfort, and improves cooperation in young or restless patients.

The authors are candid about the limitations that must be addressed before the method becomes standard practice. The study was conducted at a single center and enrolled a nearly homogeneous population of Han children from central China, raising questions about generalizability to other ethnic and regional groups, particularly given evidence that the historical Greulich-Pyle atlas itself fits contemporary Asian children imperfectly. Adolescents with already closed epiphyses were excluded, and in children aged zero to two years the ossification centers are often absent or extremely tiny, which markedly reduces ultrasound measurement accuracy in that youngest group. Ultrasound is also inherently operator-dependent, and although standardized scan planes and numeric ratios mitigate variability, multi-center and multi-operator validation studies remain necessary. Even so, the combination of a 0.98 correlation, 0.99 reliability, zero radiation, and a single-limb scanning protocol makes this summed ossification ratio one of the most compelling candidates yet for bringing bone age assessment out of the X-ray room and into the pediatric clinic.

Subject of Research: A radiation-free ultrasound parameter for pediatric bone age assessment based on summed ossification ratios

Article Title: Summation of ossification ratios of the distal radius, distal ulna, and proximal radius: a novel parameter for pediatric bone age assessment

Article References: Zeng, F., Shen, X., Xiao, X., & Wu, X. (2026). Summation of ossification ratios of the distal radius, distal ulna, and proximal radius: a novel parameter for pediatric bone age assessment. Pediatric Radiology. https://doi.org/10.1007/s00247-026-06772-2

Image Credits: AI Generated

DOI: 10.1007/s00247-026-06772-2

Keywords: bone age, ultrasound, ossification ratio, pediatrics, Greulich-Pyle atlas, radiation-free imaging, distal radius, proximal radius, skeletal maturity, pediatric radiology, growth disorders, diagnostic imaging