Automated age estimation of young individuals based on 3D knee MRI using deep learning
- PDF / 6,554,266 Bytes
- 15 Pages / 595.224 x 790.955 pts Page_size
- 9 Downloads / 237 Views
ORIGINAL ARTICLE
Automated age estimation of young individuals based on 3D knee MRI using deep learning Markus Auf der Mauer1 · Eilin Jopp-van Well2 · Jochen Herrmann3 · Michael Groth3 · Michael M. Morlock4 · 1 ¨ Rainer Maas5 · Dennis Saring Received: 23 March 2020 / Accepted: 9 November 2020 © The Author(s) 2020
Abstract Age estimation is a crucial element of forensic medicine to assess the chronological age of living individuals without or lacking valid legal documentation. Methods used in practice are labor-intensive, subjective, and frequently comprise radiation exposure. Recently, also non-invasive methods using magnetic resonance imaging (MRI) have evaluated and confirmed a correlation between growth plate ossification in long bones and the chronological age of young subjects. However, automated and user-independent approaches are required to perform reliable assessments on large datasets. The aim of this study was to develop a fully automated and computer-based method for age estimation based on 3D knee MRIs using machine learning. The proposed solution is based on three parts: image-preprocessing, bone segmentation, and age estimation. A total of 185 coronal and 404 sagittal MR volumes from Caucasian male subjects in the age range of 13 and 21 years were available. The best result of the fivefold cross-validation was a mean absolute error of 0.67 ± 0.49 years in age regression and an accuracy of 90.9%, a sensitivity of 88.6%, and a specificity of 94.2% in classification (18-year age limit) using a combination of convolutional neural networks and tree-based machine learning algorithms. The potential of deep learning for age estimation is reflected in the results and can be further improved if it is trained on even larger and more diverse datasets. Keywords Age estimation · Magnetic resonance imaging · Knee · Machine learning · Convolutional neural networks
Introduction The determination of certain age limits plays a crucial role in asylum applications, criminal proceedings, and professional youth sport. It can have important consequences for the persons in question. For example, special benefits apply for underage refugees [1, 2], specific laws are enforced to accused subjects [3], or exclusion from tournaments can occur for young athletes [4–6]. The retrieval of the chronological age is required whenever there is a lack of documentation or doubt about the alleged age [7–9]. The European Asylum Support Office (EASO) has presented guidelines on age estimation methods [10]. Markus Auf der Mauer and Eilin Jopp-van Well shared first authorship. Markus Auf der Mauer
[email protected]
Extended author information available on the last page of the article.
The recommendation is to perform the assessment of the chronological age using first non-medical and then medical methods. However, the results from non-medical methods, such as personal interviews or psychological assessments, are often inconclusive [11]. Hence, medical methods are necessary. These are based on the visual inspection of growth
Data Loading...
