西藏自治区战略意义重要，藏族儿童的正常生长发育决定了藏区未来的健康状况和发展潜能。骨龄是国际公认的准确反应儿童生长发育的重要指标，但藏族儿童骨龄发育数据缺乏，主要原因之一是藏区缺乏骨龄相关专业人员及技能。AI为以有限医疗资源实现高效准确的骨龄评估提供了可能。课题组前期研发出了中国平原地区儿童的AI骨龄系统，本研究拟在此基础上，以援藏工作为契机，深入整合深度学习AI技术，通过深度CNN完成藏族儿童骨龄图像的像素池化、卷积化，使用ResNet自动提取图像高维特征进行智能分割和定量评级，开发出适合藏族儿童的新型AI骨龄系统；并用此系统深入剖析上千例藏族儿童骨龄图像，以中国儿童生长发育大数据为参考，绘制藏族儿童骨龄及生长发育曲线。研究目标的实现可填补藏区未能常规开展骨龄检查的短板，揭示藏族儿童生长发育规律，为在高寒高海拔地区及边远牧区实现大规模调查，制定健康西藏相关政策提供重要技术及数据支持。

Bone age (BA), which evaluates skeletal maturity from the radiographs of the left hand and wrist, is a crucial indicator for revealing the growth and development of children. In high plateau areas, genetic factors, special natural environment and habitat may cause Tibetan children growing slower compared to Han children from flat areas. Nevertheless, the data regarding the BA and development of Tibetan children is scarce due to lack of bone age examination and specialized doctors..Two methods are mainly used to assess BA: the Greulich and Pyle (GP) and Tanner-Whitehouse (TW3); of these, the GP atlas is generally accepted as a faster and simpler method and thus widely applied in clinical practice. However, manual assessment of BA completely depends on the reviewers’ experience to determine BA, thereby causing significant intra- and inter-observer variations. Furthermore, constant time and effort are needed to train clinical reviewers; consequently, primary and rural hospitals face a daunting task to carry out this important examination..Artificial intelligence (AI), which has high potential in reducing labor requirement and intra- and inter-observer variations, is gaining popularity in medical field, especially in radiology. Deep learning, one of the advanced AI techniques, which can automatically learn features from images, has become a hot spot in recent years..In addition to some traditional learning-based approaches, several preliminary deep learning-based BA systems have been developed using a standard database or radiographs from one or two medical centers in North America and Korea. However, these deep learning-based systems were developed for Western and Korean populations; thus, they might not be suitable for Chinese children. Moreover, the number of tested patients was relatively small and only with limited types of diseases. Therefore, these AI BA systems cannot be directly applied to populations of different ethnicities..Since 2017, a fully automated AI BA system for Chinese children based on the GP method has been developed by using 8000 BA radiographs from five medical centers nationwide in China. In 2019, we evaluated the performance of the AI system by using 745 radiographs of patients with abnormal growth and development from Peking Union Medical College Hospital. Compared to the interpretation results of experienced human reviewers, the overall BA accuracy of AI within 1 year was nearly 85%. According to the results of RMSE, MAD, and 95% limits of agreement, the degree of dispersion was also acceptable, indicating that the developed automated AI system could achieve comparable BA results to experienced reviewers for Chinese children..Herein, hundreds of left hand radiographs of Tibetan children are input into this AI module, and multiple hyperparameters are tuned in order to develop a novel fully automated AI BA system for Tibetan children. At the same time, native technicians and doctors are trained to perform bone age examination and assessment in clinical practice. Ultimately, thousands of BA images collected during the research period are analyzed by the novel AI BA system to capture the features of growing features of Tibetan Children in high plateau..This study will develop an AI BA system for Tibetan children, and popularize BA assessment in Tibet effectively and efficiently. The developmental characteristics of Tibetan children will be disclosed through large sample analysis. Most importantly, this study will provide important technical support and data basis for the Healthy Tibet Plan in China.