KMID : 0311120220630070683
|
|
Yonsei Medical Journal 2022 Volume.63 No. 7 p.683 ~ p.691
|
|
Re-Assessment of Applicability of Greulich and Pyle-Based Bone Age to Korean Children Using Manual and Deep Learning-Based Automated Method
|
|
Hwang Ji-Sun
Yoon Hee-Mang Hwang Jae-Yeon Kim Pyeong-Hwa Bak Bo-Ram Bae Byeong-Uk Sung Jin-Kyeong Kim Hwa-Jung Jung Ah-Young Cho Young-Ah Lee Jin-Seong
|
|
Abstract
|
|
|
Purpose: To evaluate the applicability of Greulich-Pyle (GP) standards to bone age (BA) assessment in healthy Korean childrenusing manual and deep learning-based methods.
Materials and Methods: We collected 485 hand radiographs of healthy children aged 2?17 years (262 boys) between 2008 and2017. Based on GP method, BA was assessed manually by two radiologists and automatically by two deep learning-based BA as sessment (DLBAA), which estimated GP-assigned (original model) and optimal (modified model) BAs. Estimated BA was com pared to chronological age (CA) using intraclass correlation (ICC), Bland-Altman analysis, linear regression, mean absolute error,and root mean square error. The proportion of children showing a difference >12 months between the estimated BA and CA wascalculated.
Results: CA and all estimated BA showed excellent agreement (ICC ¡Ã0.978, p<0.001) and significant positive linear correlations (R2¡Ã0.935, p<0.001). The estimated BA of all methods showed systematic bias and tended to be lower than CA in younger patients, andhigher than CA in older patients (regression slopes ¡Â-0.11, p<0.001). The mean absolute error of radiologist 1, radiologist 2, origi nal, and modified DLBAA models were 13.09, 13.12, 11.52, and 11.31 months, respectively. The difference between estimatedBA and CA was >12 months in 44.3%, 44.5%, 39.2%, and 36.1% for radiologist 1, radiologist 2, original, and modified DLBAAmodels, respectively.
Conclusion: Contemporary healthy Korean children showed different rates of skeletal development than GP standard-BA, andsystemic bias should be considered when determining children¡¯s skeletal maturation.
|
|
KEYWORD
|
|
Age determination by skeleton, radiography, hand bones, child, deep learning
|
|
FullTexts / Linksout information
|
|
|
|
Listed journal information
|
|
|
|