法医学杂志

法医学杂志

• •    

基于多种深度卷积神经网络模型的汉族青少年儿童肘关节X线骨龄推断

李丹阳1,2,3(), 周慧明3,4, 万雷3, 刘太昂5, 李远喆5, 汪茂文3, 王亚辉3()   

  1. 1.山西医科大学医学科学院，山西 太原 030000
    2.山西医科大学公共卫生学院，山西 太原 030000
    3.司法鉴定科学研究院 上海市法医学重点实验室 司法部司法鉴定重点实验室 上海市司法鉴定专业技术服务平台，上海 200063
    4.山西医科大学法医学院，山西 晋中 030600
    5.上海数之微信息科技有限公司，上海 200444
  • 收稿日期:2024-12-25
  • 通讯作者: 王亚辉
  • 作者简介:李丹阳（1999—），女，硕士研究生，主要从事公共卫生、法医临床学和法医人类学研究；E-mail：lidanyang19990304@163.com
  • 基金资助:
    国家重点研发计划资助项目(2022YFC3302004);国家自然科学基金资助项目(81571859);上海市2019年度“科技创新行动计划”技术标准项目(19DZ2201300);上海市法医学重点实验室资助项目(21DZ2270800);上海市司法鉴定专业技术服务平台资助项目;司法部司法鉴定重点实验室资助项目

Bone Age Estimation of Chinese Han Adolescents and Children’s Elbow Joint X-rays Based on Multiple Deep Convolutional Neural Network Models

Dan-yang LI1,2,3(), Hui-ming ZHOU3,4, Lei WAN3, Tai-ang LIU5, Yuan-zhe LI5, Mao-wen WANG3, Ya-hui WANG3()   

  1. 1.Academy of Medical Sciences, Shanxi Medical University, Taiyuan 030000, China
    2.School of Public Health, Shanxi Medical University, Taiyuan 030000, China
    3.Shanghai Key Laboratory of Forensic Medicine, Key Laboratory of Forensic Science, Ministry of Justice, Shanghai Forensic Service Platform, Academy of Forensic Science, Shanghai 200063, China
    4.School of Forensic Medicine, Shanxi Medical University, Jinzhong 030600, Shanxi Province, China
    5.Shanghai Shuzhiwei Information Technology Co. , Ltd, Shanghai 200444, China
  • Received:2024-12-25
  • Contact: Ya-hui WANG

PDF (PC)

10

补充材料

1

摘要：

目的 探讨适用于我国汉族青少年儿童肘关节X线图像的深度学习骨龄自动推断模型，并评估其性能。 方法 采集我国华东、华南、华中、西北地区6.00~<16.00周岁汉族青少年儿童肘关节正位X线图像943例（男性517例，女性426例），采用3种实验方案（方案一：将预处理后的上述图像直接输入回归模型；方案二：以“肘关节重点骨骼标注”作为标签训练分割网络，将分割后的图像输入回归模型；方案三：以“肘关节全部骨骼标注”作为标签训练分割网络，将分割后的图像输入回归模型）进行肘关节X线骨龄预测。针对分割任务，从U-Net、UNet++和TransUNet中遴选出最优网络模型作为分割网络；针对回归任务，选择VGG16、VGG19、InceptionV2、InceptionV3、ResNet34、ResNet50、ResNet101和DenseNet121进行骨龄预测。采用随机抽样的方法抽取80%样本（754例）作为训练集和验证集，用于模型拟合和超参数的调整；20%（189例）作为内部测试集，用于测试训练后模型性能。另采集104例同源6.00~<16.00周岁汉族青少年儿童肘关节正位X线图像作为外部测试集。通过比较模型预测年龄与真实生活年龄之间的平均绝对误差（mean absolute error，MAE）、均方根误差（root mean square error，RMSE）、±0.7岁的准确率（P±0.7岁）、±1.0岁的准确率（P±1.0岁），并绘制雷达图、散点图、热力图评估模型的性能。 结果 按照方案三的方法进行分割时，UNet++模型在学习率为0.000 1时的分割损失为0.000 4，准确率为93.8%，模型分割性能优异。在内部测试集中，DenseNet121模型采用该分割方法的模型预测结果最优，MAE、P±0.7岁、P±1.0岁分别为0.83岁、70.03%、84.30%。在外部测试集中，DenseNet121模型采用方案三的结果最优，平均MAE为0.89岁、平均RMSE为1.00岁。 结论 对青少年儿童肘关节X线图像进行骨龄自动推断时，在分割网络的选择上推荐使用UNet++模型，DenseNet121模型在采用方案三时的性能最优。使用分割网络，特别是以包括肱骨远端、桡骨近端、尺骨近端全部肘关节作为标注区域的分割网络能提高肘关节X线骨龄推断的准确性。

关键词: 法医人类学, 年龄推断, X线图像, 肘关节, 深度卷积神经网络, 分割网络, 汉族, 青少年, 儿童

Abstract:

Objective To explore a deep learning-based automatic bone age estimation model for elbow joint X-ray images of Chinese Han children and adolescents and evaluate its performance. Methods A total of 943 （517 males and 426 females） elbow joint X-ray images of Chinese Han children and adolescents aged 6.00 to <16.00 years were collected from East, South, Central, and Northwest China. Three experimental schemes were adopted for bone age estimation. Scheme 1: Directly input preprocessed images into a regression model; Scheme 2: Train a segmentation network using “key elbow bone annotations” as labels, then input segmented images into the regression model; Scheme 3: Train a segmentation network using “full elbow bone annotations” as labels, then input segmented images into the regression model. For segmentation, the optimal model was selected from U-Net, UNet++ and TransUNet. For regression, VGG16, VGG19, InceptionV2, InceptionV3, ResNet34, ResNet50, ResNet101, and DenseNet121 were selected for bone age estimation. The dataset was randomly split into 80% （754 images） for training/validation for model fitting and hyperparameter tuning and 20% （189 images） as an internal test set to test the performance of the trained model. An additional 104 elbow X-ray images from the same demographic and age group were collected and used as an external test set. Model performance was evaluated by comparing the mean absolute error （MAE）, root mean square error （RMSE）, accuracy within ±0.7 years （P±0.7 years） and ±1.0 years （P±1.0 years） between the estimated age and the actual age, and by drawing radar charts, scatter plots, and heatmaps. Results When segmented with “full elbow bone annotations,” the UNet++ model achieved a segmentation loss of 0.0004 and an accuracy of 93.8% at a learning rate of 0.0001. In the internal test set, DenseNet121 model with Scheme 3 yielded the best results with MAE, P±0.7 years and P±1.0 years being 0.83 years, 70.03%, and 84.30%, respectively. In the external test set, the DenseNet121 with Scheme 3 also performed best, with an average MAE of 0.89 years and an average RMSE of 1.00 years. Conclusion When performing automatic bone age estimation using elbow X-rays in Chinese Han children and adolescents, it is recommended to use the UNet for segmentation. DenseNet121 with Scheme 3 achieves optimal performance. Using a segmentation network, especially one that includes the full elbow joint including the distal humerus, proximal radius, and proximal ulna as the annotation area, can improve the accuracy of bone age estimation based on elbow X-ray images.

Key words: forensic anthropology, age estimation, X-ray image, elbow joint, deep convolutional neural network, segmentation network, Han nationality, adolescents, children

中图分类号: