The value of AI-assisted multiple bone age assessment methods for evaluating efficacy in idiopathic short stature: A retrospective study of traditional Chinese medicine treatment for girls in China.
Authors
Affiliations (2)
Affiliations (2)
- Department of Radiology, Ningbo Municipal Hospital of Traditional Chinese Medicine (TCM), Affiliated Hospital of Zhejiang Chinese Medical University, Ningbo, Zhejiang, China.
- Department of Radiology, Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China.
Abstract
There are few studies comparing the impact of different bone age (BA) assessment methods on treatment follow-up cycles. This study aimed to analyze the value of artificial intelligence-assisted multiple BA assessment methods in evaluating the efficacy of idiopathic short stature (ISS) in traditional Chinese medicine treatment. We selected 388 cases of healthy children's BA data from November 2021 to August 2024 in a single center as the control group, and 190 cases of female pediatric patients with ISS admitted in the same period as the treatment group, which were divided into 120 cases in the effective group and 70 cases in the ineffective group according to the final therapeutic efficacy. They were divided into 3 periods of 6, 9, and 12 months according to the interval of follow-up. All the children underwent left hand X-ray examinations before treatment and after follow-up, all the X-rays were taken using artificial intelligence-assisted Greulich and Pyle atlas (G&P), Tanner Whitehouse 3 (TW3), and Tanner Whitehouse China 05 (TW C) methods for assessment. We analyzed the initial and follow-up BA assessment results and their differences and compared the relationship between the BA difference and the efficacy of each method. The results showed that the overall difference in BA follow-up was significantly higher in the treatment-effective group than in the ineffective group (P < .01). The assessment value of BA difference based on TW3 and TW C methods was better than that of the G&P method in the follow-up of the 6th, 9th and 12th months (P < .01). The TW C and TW3 methods could evaluate the efficacy of the treatment early in the 6-month follow-up, while the G&P method had a diagnostic value in the 12-month follow-up. Both TW3 and TW C methods of BA assessment can be used to evaluate the treatment effect as early as the 6-month follow-up period, and the TW C method had a higher diagnostic value of efficacy in Chinese girls with ISS compared with the TW3 and G&P methods.