Yuan Li1,2, Zhizhong Huang3, Xiaoai Dong2, Weibo Liang1, Hui Xue4, Lin Zhang1, Yi Zhang5, Zhenhua Deng6. 1. Department of Forensic Genetics, West China School of Preclinical and Forensic Medicine, Sichuan University, Chengdu, 610041, Sichuan, People's Republic of China. 2. Department of Forensic Pathology, West China School of Preclinical and Forensic Medicine, Sichuan University, No. three, 17 South Renmin Road, Wuhou District, Chengdu City, 610041, Sichuan, People's Republic of China. 3. College of Computer Science, Sichuan University, No.24 South Section 1, Yihuan Road, Chengdu, 610065, China. 4. Department of Radiology, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, People's Republic of China. 5. College of Computer Science, Sichuan University, No.24 South Section 1, Yihuan Road, Chengdu, 610065, China. yzhang@scu.edu.cn. 6. Department of Forensic Pathology, West China School of Preclinical and Forensic Medicine, Sichuan University, No. three, 17 South Renmin Road, Wuhou District, Chengdu City, 610041, Sichuan, People's Republic of China. fydzh63@163.com.
Abstract
PURPOSE: To develop a deep learning bone age assessment model based on pelvic radiographs for forensic age estimation and compare its performance to that of the existing cubic regression model. MATERIALS AND METHOD: A retrospective collection data of 1875 clinical pelvic radiographs between 10 and 25 years of age was obtained to develop the model. Model performance was assessed by comparing the testing results to estimated ages calculated directly using the existing cubic regression model based on ossification staging methods. The mean absolute error (MAE) and root-mean-squared error (RMSE) between the estimated ages and chronological age were calculated for both models. RESULTS: For all test samples (between 10 and 25 years old), the mean MAE and RMSE between the automatic estimates using the proposed deep learning model and the reference standard were 0.94 and 1.30 years, respectively. For the test samples comparable to those of the existing cubic regression model (between 14 and 22 years old), the mean MAE and RMSE for the deep learning model were 0.89 and 1.21 years, respectively. For the existing cubic regression model, the mean MAE and RMSE were 1.05 and 1.61 years, respectively. CONCLUSION: The deep learning convolutional neural network model achieves performance on par with the existing cubic regression model, demonstrating predictive ability capable of automated skeletal bone assessment based on pelvic radiographic images. KEY POINTS: • The pelvis has considerable value in determining the bone age. • Deep learning can be used to create an automated bone age assessment model based on pelvic radiographs. • The deep learning convolutional neural network model achieves performance on par with the existing cubic regression model.
PURPOSE: To develop a deep learning bone age assessment model based on pelvic radiographs for forensic age estimation and compare its performance to that of the existing cubic regression model. MATERIALS AND METHOD: A retrospective collection data of 1875 clinical pelvic radiographs between 10 and 25 years of age was obtained to develop the model. Model performance was assessed by comparing the testing results to estimated ages calculated directly using the existing cubic regression model based on ossification staging methods. The mean absolute error (MAE) and root-mean-squared error (RMSE) between the estimated ages and chronological age were calculated for both models. RESULTS: For all test samples (between 10 and 25 years old), the mean MAE and RMSE between the automatic estimates using the proposed deep learning model and the reference standard were 0.94 and 1.30 years, respectively. For the test samples comparable to those of the existing cubic regression model (between 14 and 22 years old), the mean MAE and RMSE for the deep learning model were 0.89 and 1.21 years, respectively. For the existing cubic regression model, the mean MAE and RMSE were 1.05 and 1.61 years, respectively. CONCLUSION: The deep learning convolutional neural network model achieves performance on par with the existing cubic regression model, demonstrating predictive ability capable of automated skeletal bone assessment based on pelvic radiographic images. KEY POINTS: • The pelvis has considerable value in determining the bone age. • Deep learning can be used to create an automated bone age assessment model based on pelvic radiographs. • The deep learning convolutional neural network model achieves performance on par with the existing cubic regression model.
Entities:
Keywords:
Age determination by skeleton; Forensic anthropology; Machine learning; Pelvis; Radiography
Authors: Nicolene Lottering; Clair L Alston-Knox; Donna M MacGregor; Maree T Izatt; Caroline A Grant; Clayton J Adam; Laura S Gregory Journal: J Forensic Sci Date: 2016-11-25 Impact factor: 1.832
Authors: Daniel Wittschieber; Andreas Schmeling; Sven Schmidt; Walter Heindel; Heidi Pfeiffer; Volker Vieth Journal: Forensic Sci Med Pathol Date: 2012-11-09 Impact factor: 2.007
Authors: Daniel Wittschieber; Volker Vieth; Christoph Domnick; Heidi Pfeiffer; Andreas Schmeling Journal: Int J Legal Med Date: 2012-10-02 Impact factor: 2.686
Authors: Vilma Pinchi; Federica De Luca; Martina Focardi; Francesco Pradella; Giulia Vitale; Federico Ricciardi; Gian-Aristide Norelli Journal: Leg Med (Tokyo) Date: 2016-04-26 Impact factor: 1.376
Authors: Li-Qin Peng; Yu-Cheng Guo; Lei Wan; Tai-Ang Liu; Peng Wang; Hu Zhao; Ya-Hui Wang Journal: Int J Legal Med Date: 2022-01-18 Impact factor: 2.686
Authors: Sebastian Gassenmaier; Juergen F Schaefer; Konstantin Nikolaou; Michael Esser; Ilias Tsiflikas Journal: Eur Radiol Date: 2020-07-28 Impact factor: 5.315
Authors: Carlos A Peña-Solórzano; David W Albrecht; Richard B Bassed; Michael D Burke; Matthew R Dimmock Journal: Forensic Sci Int Date: 2020-10-18 Impact factor: 2.395
Authors: Markus Auf der Mauer; Eilin Jopp-van Well; Jochen Herrmann; Michael Groth; Michael M Morlock; Rainer Maas; Dennis Säring Journal: Int J Legal Med Date: 2020-12-17 Impact factor: 2.686