Construction of a Simplified Bone Age Assessment Model in Chinese Children Aged 3 to 18 Years
-
By
-
Hua-hong Wu
-
Ya-qin Zhang
-
Cheng-dong Yu
-
Yang Li
-
Wen Shu
-
Tao Li
-
Gui-min Huang
-
Dong-qing Hou
-
Fang-fang Chen
-
Jun-ting Liu
-
Shao-li Li
-
Xin-nan Zong
-
January 15, 2025
-
Clinical Scorecard: Development of an Enhanced Model for Bone Age Evaluation in Chinese Children Aged 3 to 18 Years
At a Glance
| Category | Detail |
| Condition | Bone age assessment (BAA) for evaluating physiological maturity and growth status in children |
| Key Mechanisms | Simplified bone age evaluation using linear regression models based on grades of radius, ulna, and metacarpal bones from left-hand X-rays |
| Target Population | Chinese children aged 3 to 18 years |
| Care Setting | Pediatric endocrinology and general pediatric clinical settings |
Key Highlights
- Traditional BAA methods are complex, time-consuming, and require professional expertise.
- A simplified 3-bone model (radius, ulna, metacarpal I) achieves high accuracy (R² > 0.94, RMSE < 0.5 years).
- Incorporating pubertal stage improves bone age assessment accuracy.
Guideline-Based Recommendations
Diagnosis
- Use the Tanner-Whitehouse 3-China radius-ulna-short bone (TW3-C-RUS) method as the gold standard for bone age assessment.
- Perform bone age assessment using left-hand wrist radiographs.
- Group children by pubertal stage to refine assessment accuracy.
Management
- Apply simplified 3-bone regression models for rapid bone age evaluation in clinical practice.
- Consider bone grade consistency to use single or few bones as proxies when applicable.
- Use regression models when bone grades are inconsistent.
Monitoring & Follow-up
- Conduct reassessments by trained pediatricians to minimize subjective bias, ensuring discrepancies ≤ 0.5 years.
- Calibrate X-ray devices daily to maintain consistency in imaging results.
Risks
- Potential inaccuracies due to individual variations in developmental patterns.
- Early fusion of carpal bones limits their use in older children.
- Overfitting and information loss may occur when converting bone grades to scores and mapping to bone age.
Patient & Prescribing Data
Normal Chinese children aged 3 to 18 years undergoing bone age assessment.
Simplified models enable quicker and accurate bone age evaluation, facilitating timely clinical decisions for growth-related conditions.
Clinical Best Practices
- Train nonmedical personnel thoroughly for preliminary bone age assessments to improve efficiency.
- Use multiple rounds of assessment and blinded review to reduce subjective bias.
- Incorporate pubertal staging in bone age evaluation models for enhanced precision.
- Prefer the TW3-C-RUS method over carpal bone assessment in older children due to early carpal fusion.
- Utilize regression models directly converting bone grades to bone age to reduce information loss.
References