Applicability of Greulich–Pyle and Tanner–Whitehouse grading methods to MRI when assessing hand bone age in forensic age estimation: A pilot study

• Published in: Forensic science international Vol. 266; pp. 281 - 288
• Main Authors: Urschler, Martin, Krauskopf, Astrid, Widek, Thomas, Sorantin, Erich, Ehammer, Thomas, Borkenstein, Martin, Yen, Kathrin, Scheurer, Eva
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.09.2016; Elsevier Limited
• Subjects: Accuracy; Age; Age determination; bone formation; Bone surgery; Bones; Chronology; Forensic age estimation; Forensic engineering; Forensic medicine; Forensic osteology; Forensic science; Forensic sciences; Gender; Grading; Greulich–Pyle; Hand (anatomy); Hand–wrist; Hypotheses; Identification documents; Ionizing radiation; Magnetic resonance imaging; Methods; Morphology; MRI; NMR; Nuclear magnetic resonance; Ossification; Pathology; people; Radiographs; Radiography; Regression analysis; skeletal development; Studies; Tanner–Whitehouse; X-ray; Greulich–Pyle; Hand–wrist; X-ray; Tanner–Whitehouse; MRI; Forensic age estimation
• ISSN: 0379-0738; 1872-6283
• DOI: 10.1016/j.forsciint.2016.06.016

•Forensic age estimation with Greulich–Pyle and Tanner–Whitehouse shown on MRI data.•For the first time a comparison of GP and TW2 on MRI and X-ray data is performed.•High inter-rater agreement indicates feasibility of applying GP and TW2 to MRI data. Determination of skeletal development is a key pillar in forensic age estimation of living persons. Radiological assessment of hand bone age is widely used until the age of about 17–18 years, applying visual grading techniques to hand radiographs. This study investigated whether Greulich–Pyle (GP) and Tanner–Whitehouse (TW2) grading can be equally used for magnetic resonance imaging (MRI) data, which would offer the huge benefit of avoiding ionizing radiation. In 18 subjects aged between 7 and 17 years a radiograph and an MRI scan of the hand were performed. Epiphyseal ossification of hand bones was rated by two blinded radiologists with both GP and TW2. Correlation between hand MRIs and radiographs was analyzed by linear regression and inter-observer agreement was assessed. Correlation between age estimates from MRI and radiographs was high for both GP (r2=0.98) and TW2 (r2=0.93). MRI showed a tendency to estimate age slightly lower for 14–18 year-olds, which would be favorable regarding majority age determination in case this result could be reproduced using a currently not existing reference estimation method based on MRI data. Inter-observer agreement was similar for GP in radiographs and MRI, while for TW2, agreement in MRI was lower than in radiographs. In spite of limitations regarding sample size and recruited subjects, our results indicate that the use of GP and TW2 on MRI data offers the possibility of hand bone age estimation without the need for ionizing radiation.