基于深度学习的儿童骨盆X线片质量控制模型的构建与应用  

Development and Application of Deep Learning-Based Model for Quality Control of Children Pelvic X-Ray Images

作　　者：刘峙辰 林锦聪 谢坤杰 沙佳 陈旭 雷伟 黄鲁豫 严亚波 LIU Zhichen;LIN Jincong;XIE Kunjie;SHA Jia;CHEN Xu;LEI Wei;HUANG Luyu;YAN Yabo(Department of Orthopaedics,Xijing Hospital,Air Force Medical University,Xi'an,710032)

机构地区：[1]空军军医大学西京医院骨科,西安市710032

出　　处：《中国医疗器械杂志》2024年第2期144-149,共6页Chinese Journal of Medical Instrumentation

摘　　要：目的提出一种基于深度学习的儿童骨盆正位X线片质量评估方法,构建诊断模型并验证其临床可行性。方法回顾性收集3247例儿童骨盆正位X线片,随机分为训练数据集、验证数据集及测试数据集。构建人工智能模型,评估质量控制模型可靠性。结果模型的诊断准确率、ROC曲线下面积、灵敏度及特异度分别为99.4%、0.993、98.6%和100.0%。模型的骨盆倾斜指数95%一致性界限为-0.052~0.072;骨盆旋转指数95%一致性界限为-0.088~0.055。结论该研究首次尝试将AI算法应用于儿童骨盆X线片的质量评估,并显著改善了儿童发育性髋关节发育不良的诊疗现状。Objective A deep learning-based method for evaluating the quality of pediatric pelvic X-ray images is proposed to construct a diagnostic model and verify its clinical feasibility.Methods Three thousand two hundred and forty-seven children with anteroposteric pelvic radiographs are retrospectively collected and randomly divided into training datasets,validation datasets and test datasets.Artificial intelligence model is conducted to evaluate the reliability of quality control model.Results The diagnostic accuracy,area under ROC curve,sensitivity and specificity of the model are 99.4%,0.993,98.6%and 100.0%,respectively.The 95%consistency limit of the pelvic tilt index of the model is-0.052-0.072.The 95%consistency threshold of pelvic rotation index is-0.088-0.055.Conclusion This is the first attempt to apply AI algorithm to the quality assessment of children's pelvic radiographs,and has significantly improved the diagnosis and treatment status of DDH in children.

关 键 词：发育性髋关节发育不良 骨盆X线片 人工智能 软件 

分 类 号：R197.39[医药卫生—卫生事业管理] R445[医药卫生—公共卫生与预防医学]