ABSTRACT
BACKGROUND: In China, three bone age assessment methods have been widely used in the medical and sports fields, including the Greulich-Pyle atlas method (GP method), CHN scoring method (CHN method), and China 05 method. A large-sample empirical study is required to determine which method is more suitable for assessing bone age of children and adolescents. OBJECTIVE: To provide a scientific evidence for appropriate bone age evaluation standards for children and adolescents in the eastern developed areas, by comparing the GGP method, CHN method and China 05 method based on samples of healthy children from Shanghai. METHODS: A total of 4 152 healthy children and adolescents (2 185 boys and 1 967 girls) from the urban area of Shanghai were selected for the study. Their digital X-ray of the left hand and wrist were collected and evaluated by the GGP method, CHN method and China 05 method. The difference between the bone age and the chronological age was used to assess the applicability of different bone age standards. The study was approved by the Ethics Committee of Shanghai Research Institute of Sports Science, and informed consent was given by all parents of the enrolled students. RESULTS AND CONCLUSION: For the GP method, the difference between bone age and chronological age in both genders at the age of ≥ 8 years was-0.12 to-0.65 year with significant difference, except for 8-year-old girls. The significant age difference at the age of ≥ 9 years was 0.18 to 1.62 year, except for the 9-year-old age group. For the CHN method, the difference between bone age and chronological age among 6-17-year-old boys and 6-16-year-old girls was 0.42 to 1.56 years (P 0.05), and-0.60 in 18-year-old boys (P < 0.01); the age difference among 6-17-year-old girls was-0.01 to 0.56 year, and the difference was not significant in most age groups. Among the three methods, the result of China 05 method is relatively better, which is the best method that matches the current development of teenagers in Shanghai, suggesting that the China 05 method is more suitable for the eastern developed areas with economic level similar to Shanghai. All the three methods have some limitations. Considering the long-term growth trend of adolescents, it is necessary to revise the current bone age evaluation standards.
ABSTRACT
Objective@#To build an automatic bone age assessment system based on China 05 Bone Age Standard and the latest deep learning technology, and preliminary clinical verification was carried out.@*Methods@#The left-hand radiographs of 5 000 children with suspected metabolic disorders were acquired from Wuxi Children′s Hospital. Among these cases, 2 351 patients were randomly chosen as training set, and 101 patients were randomly used as validation set. Four professional pediatric radiologists annotated the development stage according to the China 05 RUS-CHN standard with double-blind method. The mean value of the bone age assessed by experts was the reference standard which was used to train and validate the deep learning mothods based artificial intelligence (AI) model. Accuracy, mean absolute error (MAE), root mean squared error (RMSE) and time efficiency of bone age assessment were compared by using Chi-square test and t test and F test between resident doctors and AI model in the validation set.@*Results@#The MAE and RMSE was (0.37±0.35) years and 0.50 years between AI model and reference standard, respeactively. When the error range was within ±1.0, ±0.7 and ±0.5 years, the accuracy of model on the validation set was 94.1% (95/101), 89.1% (90/101), 74.3% (75/101) respectively. The accuracy between two resident doctors and AI prediction wasn′t statistical significant (P>0.05).@*Conclusion@#The AI model of bone age assessment based on deep learning is feasible and has the characteristics of high accuracy and efficiency.