法医学杂志

法医学杂志 ›› 2020, Vol. 36 ›› Issue (5): 622-630.DOI: 10.12116/j.issn.1004-5619.2020.05.004

所属专题： 虚拟法医人类学

• 专题 • 上一篇    下一篇

运用3种卷积神经网络模型对青少年骨盆骨龄评估的比较

彭丽琴1，2， 万雷2， 汪茂文2， 李卓2， 王鹏3， 刘太昂3， 王亚辉2， 赵虎1   

  1. 1. 中山大学中山医学院法医学系 广东省法医学转化医学工程技术研究中心，广东 广州 510080；2. 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；3. 上海真谱信息科技有限公司，上海 200444
  • 收稿日期:2020-07-01 发布日期:2020-10-25 出版日期:2020-10-28
  • 通讯作者: 王亚辉，男，副研究员，硕士研究生导师，主要从事法医临床学及法医人类学研究；E-mail：wangyh@ssfjd.cn 赵虎，男，博士，教授，博士研究生导师，主要从事法医精神病学与法医临床学研究；E-mail：zhaohu3@mail.sysu.edu.cn
  • 作者简介:彭丽琴（1995—），女，土家族，硕士研究生，主要从事法医临床学研究；E-mail：pengliqinmn@163.com
  • 基金资助:
    国家自然科学基金面上资助项目（81571859，81273350，81471829）；上海市法医学重点实验室资助项目（17DZ2273200）； 上海市司法鉴定专业技术服务平台资助项目（19DZ2292700）；上海市法医学重点实验室开放基金资助项目（KF1706）

Comparison of Three CNN Models Applied in Bone Age Assessment of Pelvic Radiographs of Adolescents

PENG Li- qin1,2 , WAN Lei2 , WANG Mao- wen2 , LI Zhuo2 , WANG Peng3 , LIU Tai- ang3 , WANG Ya- hui2 , ZHAO Hu1   

  1. 1. Guangdong Province Translational Forensic Medicine Engineering Technology Research Center, Faculty of Forensic Medicine, Zhongshan School of Medicine, Sun-Yat Sen University, Guangzhou 510080, China; 2. 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; 3. Shanghai Zhenpu Information Technology Co. Ltd., Shanghai 200444, China
  • Received:2020-07-01 Online:2020-10-25 Published:2020-10-28

PDF (PC)

630

摘要： 目的 比较VGG19、Inception-V3、Inception-ResNet-V2 3种深度学习（deep learning，DL）模型基于骨盆X线片图像进行骨龄自动评估的性能。 方法 采集我国5省市11.0~<21.0周岁汉族青少年骨盆X线片图像962例（男性481例，女性481例），将上述图像进行预处理作为研究对象。采用随机抽样的方法抽取80%作为训练集、验证集，用于模型拟合和超参数的调整。20%作为测试集，用于评估模型泛化的能力。通过比较模型估计值与生活年龄的均方根误差（root mean square error，RMSE）、平均绝对误差（mean absolute error，MAE）以及绘制Bland-Altman散点图来评估3种模型的性能。 结果 VGG19模型预测年龄与生活年龄的平均RMSE、MAE分别为1.29、1.02岁，Inception-V3模型预测年龄与生活年龄的平均RMSE、MAE分别为1.17、0.82岁，Inception-ResNet-V2模型预测年龄与生活年龄的平均RMSE、MAE分别为1.11、0.84岁。Bland-Altman散点图显示Inception-ResNet-V2模型的差值的均值最小。 结论 在对青少年骨盆的自动骨龄评估中，Inception-ResNet-V2模型性能最优，Inception-V3模型与VGG19模型性能相当。

关键词: 法医人类学；年龄测定, 骨骼；骨盆；图像识别；深度学习；卷积神经网络；汉族；青少年

Abstract: Objective To compare the performance of three deep-learning models （VGG19, Inception-V3 and Inception-ResNet-V2） in automatic bone age assessment based on pelvic X-ray radiographs. Methods A total of 962 pelvic X ray radiographs taken from adolescents （481 males, 481 females） aged from 11.0 to 21.0 years in five provinces and cities of China were collected, preprocessed and used as objects of study. Eighty percent of these X ray radiographs were divided into training set and validation set with random sampling method and used for model fitting and hyper-parameters adjustment. Twenty percent were used as test sets, to evaluate the ability of model generalization. The performances of the three models were assessed by comparing the root mean square error （RMSE）, mean absolute error （MAE） and Bland-Altman plots between the model estimates and the chronological ages. Results The mean RMSE and MAE between bone age estimates of the VGG19 model and the chronological ages were 1.29 and 1.02 years, respectively. The mean RMSE and MAE between bone age estimates of the Inception-V3 model and the chronological ages were 1.17 and 0.82 years, respectively. The mean RMSE and MAE between bone age estimates of the Inception-ResNet-V2 model and the chronological ages were 1.11 and 0.84 years, respectively. The Bland-Altman plots showed that the mean value of differences between bone age estimates of Inception-ResNet-V2 model and the chronological ages was the lowest. Conclusion In the automatic bone age assessment of adolescent pelvis, the Inception-ResNet-V2 model performs the best while the Inception-V3 model achieves a similar accuracy as VGG19 model.

Key words: forensic anthropology, age determination by skeleton, pelvis, image recognition, deep learning, convolutional neural networks, Han nationality, adolescents

中图分类号: