Age Assessment of Youth and Young Adults Using Magnetic Resonance Imaging of the Knee: A Deep Learning Approach

Abstract

BackgroundBone age assessment (BAA) is an important tool for diagnosis and in determining the time of treatment in a number of pediatric clinical scenarios, as well as in legal settings where it is used to estimate the chronological age of an individual where valid documents are lacking. Traditional methods for BAA suffer from drawbacks, such as exposing juveniles to radiation, intra- and interrater variability, and the time spent on the assessment. The employment of automated methods such as deep learning and the use of magnetic resonance imaging (MRI) can address these drawbacks and improve the assessment of age.ObjectiveThe aim of this paper is to propose an automated approach for age assessment of youth and young adults in the age range when the length growth ceases and growth zones are closed (14-21 years of age) by employing deep learning using MRI of the knee.MethodsThis study carried out MRI examinations of the knee of 402 volunteer subjects—221 males (55.0%) and 181 (45.0%) females—aged 14-21 years. The method comprised two convolutional neural network (CNN) models: the first one selected the most informative images of an MRI sequence, concerning age-assessment purposes; these were then used in the second module, which was responsible for the age estimation. Different CNN architectures were tested, both training from scratch and employing transfer learning.ResultsThe CNN architecture that provided the best results was GoogLeNet pretrained on the ImageNet database. The proposed method was able to assess the age of male subjects in the range of 14-20.5 years, with a mean absolute error (MAE) of 0.793 years, and of female subjects in the range of 14-19.5 years, with an MAE of 0.988 years. Regarding the classification of minors—with the threshold of 18 years of age—an accuracy of 98.1% for male subjects and 95.0% for female subjects was achieved.ConclusionsThe proposed method was able to assess the age of youth and young adults from 14 to 20.5 years of age for male subjects and 14 to 19.5 years of age for female subjects in a fully automated manner, without the use of ionizing radiation, addressing the drawbacks of traditional methods.

Chronological Age Assessment in Young Individuals Using Bone Age Assessment Staging and Nonradiological Aspects: Machine Learning Multifactorial Approach

“…For studies that employ BA concepts to predict the CA of subjects, there are studies by Dallora et al [ 52 ] and Stern et al [ 53 ]. Both employ MRI as the medical imaging of choice, and most importantly, they are not based on traditional BAA to make their predictions of CA.…”

“…On the other hand, this problem has a reduced risk of occurring with algorithms able to analyze images pixel by pixel [ 55 ]. Dallora et al [ 52 ] used knee MRI images and achieved an MAE of 0.793 years for male subjects in the range of 14 to 20 years, and 0.988 years for female subjects in the range of 14 to 19 years. Stern et al [ 53 ] used MRI images of the hand and achieved an MAE of 0.82 years for male subjects in the range of 13 to 19 years.…”

Automated age estimation of young individuals based on 3D knee MRI using deep learning

“…In addition, the current work is dedicated to validating the results of a previous work [ 26 ] from 2018, but on a larger database, and then to developing a new robust and automated framework for age assessment based on knee MRIs. Furthermore, it can serve as a good comparison to a similar study by Dallora et al [ 30 ] published in 2019, which also used deep learning and knee MRIs. Finally, the motivation to show that the promising results of Stern et al [ 27 ]—who used deep learning for age estimation on the hand, collarbone, and teeth—are also suitable for another anatomical site, namely the knee.…”

“…Stern et al [21] pre-trained their CNN using information about the maturation of the growth plates by radiological assessment. In contrast, the models of Dallora et al [30] were pre-trained on ImageNet [62], a large database of roughly 3.2 million images of animals, vehicles, etc. Besides age regression, the studies mentioned above delivered results on the classification task as well.…”

“…In age estimation, artificial neural networks (ANNs) seem to be particularly useful because they have the ability to independently extract and learn relevant information from data for a specific task [19]. Recent studies have delivered encouraging outcomes [20][21][22][23][24][25][26][27][28][29][30].…”