Scholarlink

Background <p>The Tanner-Whitehouse 3 (TW3) method is a widely used standard for bone age (BA) assessment. Although the TW3-China (TW3-C) version was developed specifically for the Chinese population, its complexity hinders routine clinical use.</p> Objectives <p>To construct a simplified TW3-based BA assessment method using only three key bones in order to assist pediatricians in quickly assessing BA in Chinese children.</p> Methods <p>A total of 2,622 eligible healthy children aged 6–18 years were selected from a population-based cross-sectional study in Beijing (2022–2023). Left-hand wrist radiographs were collected from the participants. The TW3-C-RUS assessment of 13 bones served as the conventional reference for skeletal maturity score (SMS) and BA. Three key bones were selected by the strength of their Pearson correlation with chronological age. Three weighting protocols were designed, with the optimal weighting determined by the minimum deviation of SMS and BA from the conventional reference. Age-for-SMS percentile curves were established based on three key bones using the optimal weighting. An external validation involving 335 participants was conducted to further evaluate the agreement between the simplified and reference methods. </p> Results <p>The “radius, ulna, and metacarpal I” exhibited the strongest correlation with age (r = 0.91, 0.88, and 0.90,) and were selected as key bones. Among the three weighting protocols evaluated, the high radius weighting protocol (radius 1/2, ulna 1/4, metacarpal I 1/4) achieved the highest correlation with the reference SMS (r = 0.99) and the lowest root mean square error (33.0). Age-for-SMS percentile curves were then constructed based on the “radius-ulna-metacarpal I” combination with the high radius weighting protocol. External validation demonstrated a mean BA difference of 0.02 years (95% CI: –0.03 to 0.07) between the simplified and standard TW3-C-RUS methods, with 94.9% of estimates differing by &lt;1 year and 77.2% by &lt;0.5 year. </p> Conclusion <p>This simplified BA method demonstrated high consistency with the standard TW3-C-RUS method and may be practically useful to quick evaluate of BA in Chinese children.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Construction and Validation of a Simplified Bone Age Assessment Method Based on Three Key Bones from the TW3 system

  • 期刊论文

  • 发表日期：

    2026年4月2日

查看全文
  • Huahong Wu,
  • Yaqin Zhang,
  • Chengdong Yu,
  • Fangfang Chen,
  • Junting Liu,
  • Shaoli Li,
  • Xin’nan Zong

摘要

Background

The Tanner-Whitehouse 3 (TW3) method is a widely used standard for bone age (BA) assessment. Although the TW3-China (TW3-C) version was developed specifically for the Chinese population, its complexity hinders routine clinical use.

Objectives

To construct a simplified TW3-based BA assessment method using only three key bones in order to assist pediatricians in quickly assessing BA in Chinese children.

Methods

A total of 2,622 eligible healthy children aged 6–18 years were selected from a population-based cross-sectional study in Beijing (2022–2023). Left-hand wrist radiographs were collected from the participants. The TW3-C-RUS assessment of 13 bones served as the conventional reference for skeletal maturity score (SMS) and BA. Three key bones were selected by the strength of their Pearson correlation with chronological age. Three weighting protocols were designed, with the optimal weighting determined by the minimum deviation of SMS and BA from the conventional reference. Age-for-SMS percentile curves were established based on three key bones using the optimal weighting. An external validation involving 335 participants was conducted to further evaluate the agreement between the simplified and reference methods.

Results

The “radius, ulna, and metacarpal I” exhibited the strongest correlation with age (r = 0.91, 0.88, and 0.90,) and were selected as key bones. Among the three weighting protocols evaluated, the high radius weighting protocol (radius 1/2, ulna 1/4, metacarpal I 1/4) achieved the highest correlation with the reference SMS (r = 0.99) and the lowest root mean square error (33.0). Age-for-SMS percentile curves were then constructed based on the “radius-ulna-metacarpal I” combination with the high radius weighting protocol. External validation demonstrated a mean BA difference of 0.02 years (95% CI: –0.03 to 0.07) between the simplified and standard TW3-C-RUS methods, with 94.9% of estimates differing by <1 year and 77.2% by <0.5 year.

Conclusion

This simplified BA method demonstrated high consistency with the standard TW3-C-RUS method and may be practically useful to quick evaluate of BA in Chinese children.