Infant bone age estimation based on fibular shaft length: model development and clinical validation

Area of Science:

• Pediatric radiology
• Skeletal development
• Medical imaging analysis

Background:

• Current infant bone age estimation relies on hand/wrist or knee radiographs, with unknown accuracy and reproducibility.
• Ossification center counting is another method for infant bone age assessment.

Purpose of the Study:

• To develop and validate a novel infant bone age estimation technique utilizing fibular shaft length.
• To compare the accuracy and reproducibility of the fibular shaft length method against established conventional techniques.

Main Methods:

• Retrospective review of 247 infant skeletal surveys to develop and test a linear regression model based on fibular shaft length.
• Comparison of the proposed method with Sontag, Elgenmark, Greulich and Pyle, and Pyle and Hoerr methods using pediatric radiologists.
• Validation using lower-extremity radiographs of 114 infants.

Main Results:

• The fibular shaft length method demonstrated superior accuracy and reproducibility compared to all conventional methods.
• Testing and validation datasets showed consistent accuracy, with root-mean-square errors around 36-37 days and mean absolute errors around 28-31 days.

Conclusions:

• Fibular shaft length provides a highly accurate and reproducible method for infant bone age estimation.
• This technique presents a more reliable alternative to current conventional methods for assessing skeletal maturity in infants.