Forensic Age Determination Using MRI Scans of the Ankle: Applying Two Classifications to Assess Ossification

Abstract

Purpose

In forensic age determination, e.g. for legal proceedings, exceeded age limits may be relevant. To investigate age-related differences in skeletal development, the recommendations of the Study Group on Forensic Age Diagnostics (AGFAD) rely on imaging techniques using ionizing radiation (including orthopantomograms and radiographs of the hand). Vieth et al. and Ottow et al. have proposed MRI classifications for epi-/diaphyseal fusion of the knee joint to determine different age limits. The aim of the present study was to verify whether these two classifications could also be applied to MRI of the ankle.

Materials and Methods

MRI images of the ankle from 333 patients (160 female, 173 male) ranging in age from 10 to 28 years were retrospectively analyzed. T1-weighted turbo spin-echo (TSE) sequences and T2-weighted fat-suppressed sequences were analyzed for the two classifications. The different ossification stages of the two classifications were determined and the corresponding chronological ages were assigned. In addition, gender-specific differences were analyzed. Intra- and inter-observer variability was determined using Cohen’s kappa.

Results

With the classification of Ottow et al., the completion of the 14^th year of life could be determined in both sexes. With the classification of Vieth et al, the completion of the 14^th year of life could be determined in both sexes and the 18^th year of life in male patients. Intra-observer and inter-observer variability was very good and good, respectively (κ > 0.87 and κ > 0.72).

Conclusion

In the present study, it was also possible to use both classifications for MRI of the ankle joint. The method offers the potential of an alternative or at least supplementary radiation-free assessment criterion in forensic age estimation.

Key Points

• MRI scans of the ankle can be used for forensic age determination.

• Classifications developed for the knee joint can also be used on the ankle.

• The applied classifications based on Vieth et al. and Ottow et al. can be used as an alternative or, at the least, an additional method when determining legally relevant age limits.

Citation Format

• Wernsing MF, Malokaj V, Kunz SN et al. Forensic Age Determination Using MRI Scans of the Ankle: Applying Two Classifications to Assess Ossification. Fortschr Röntgenstr 2024; DOI 10.1055/a-2379-8785

Keywords

forensic age estimation - MRI - bone age - ossification stages

Introduction

In addition to clinical applications for questions of endocrinology and pediatrics, skeletal age determination plays an important role in forensic age diagnostics [1]. For many legal decisions, exceeding or falling below legally defined age limits is decisive. These age limits include 14, 18, and 21 years. A person under the age of 14 is considered a child in Germany and is not criminally responsible. Between the ages of 14 and 17, juvenile criminal law applies. If a person is between 18 and 21 years old, an individual decision is made as to whether juvenile or adult criminal law applies. If a person exceeds the age of 21, they are tried under adult criminal law. When a person reaches the age of 18, they have reached the age of majority, which entails legal and social responsibilities [2]. If valid identification documents are missing and dubious age information is provided, authorities and courts can request reports on forensic age diagnostics [3] [4]. As far as records indicate, there has been an increase in such reports [4]. In response to the growing demand for forensic age determination and the related discussion about proper procedures, the German Society of Legal Medicine (DGRM) founded the Working Group for Forensic Age Diagnostics (AGFAD) in 2001 [5] [6] [7]. The working group recommends a standardized procedure for forensic age diagnostics. The recommendation initially includes a medical history and physical examination. The imaging procedures used are an X-ray of the left hand and an X-ray of the teeth [8] [9] [10]. If the skeletal maturity of the hand is already complete, it is recommended to perform an additional examination of the medial clavicular epiphysis using computed tomography (CT) [11]. This approach aims to achieve a high degree of accuracy for age estimation.

The minimum age concept frequently used in forensic age determination is intended to ensure that the age estimate is in favor of the person being examined due to possible statistical deviations. The minimum age is the youngest age of a person in the reference population of a study who had the characteristic features identified. This ensures that the forensic age of the person is not overestimated but is lower than the actual age [4]. In this context, however, it must be stated critically that the validity of the currently used age determination correlates, in case of doubt, with the size and quality of the reference population. There is currently no or very inadequate comparative data, particularly for unaccompanied minor refugees.

The use of imaging procedures (X-rays and CT) for forensic age diagnostics is controversial because of the associated exposure to radiation without medical indication [12]. Imaging procedures without the use of X-rays are therefore preferable. In addition to ultrasound-assisted diagnostics, magnetic resonance imaging (MRI) has already been investigated in several studies as a possible alternative to conventional X-ray images, and it has been described as a promising method [13] [14] [15]. In addition to MRI of the hand, MRI of the knee joint has also been suggested as a possible tool for forensic age determination. There are only a few studies on forensic age diagnostics using MRI data sets of the ankle joint [16] [17] [18] [19] [20], and legally relevant age limits could only be partially determined in these studies. The classification by Vieth et al., as well as the classification by Ottow et al., was developed based on a cohort of 694 MRI scans of the knee joint [21] [22]. Using this cohort, Ottow et al. were able to determine the 14^th and 16^th year of life for both sexes, whereas the classification based on Vieth et al. was able to determine the completion of the 18^th year of life. In contrast to the knee joint, the upper ankle joint [23] is a distal joint, and it is therefore easily accessible for examination without the patient having to be placed entirely in the MR scanner, which has advantages compared to more proximal joints. The use of portable MR scanners – as used in the study by Terada et al. for determining bone age in the hand – would also be an option [24] [25].

The aim of the present study was to examine whether the classification systems for forensic age diagnostics by Ottow et al. and Vieth et al. can be transferred to MRI scans of the ankle. The focus was on determining legally relevant age limits, such as the completion of the 14^th and 18^th years of life, which could be addressed with these two classifications. In addition, the possibility of determining the age of 21, which is another important age limit, should also be investigated.

Materials and Methods

Approval of the local ethics committee to conduct the study was obtained (No. 175/20).

Patients

The study was based on a retrospective review of MRI scans of the ankle joint between the ages of 10 and 28, which were performed between October 2011 and March 2020.

The primary indication and medical question for the MRI scans were divided into 10 different categories. These are shown in [Table 1]. If a pathology hindered assessment of the epiphyseal plates, the scan was excluded. Additional exclusion criteria included implants in the area of the epiphyseal plate, as well as incomplete scans, poor image quality, or nonconforming MRI sequences that were not comparable to those of Ottow et al. or Vieth et al. As a prerequisite, we obtained consent from patients or their legal guardians for a subsequent pseudonymized evaluation.

Table 1 Indication and medical question for the MRI scans.

	Frequency	Percent
Distortion/trauma	209	62.8
Osteomyelitis	38	11.4
Osteochondrosis dissecans	41	12.3
Benign bone tumors	19	5.7
Torsion angle	7	2.1
Arthritis	7	2.1
Achilles tendonitis	5	1.5
Malignant bone tumors	3	0.9
Tendinopathy	3	0.9
Study	1	0.3
Total	333	100.0

The MRI scans were performed on a 1.5 Tesla scanner (Magnetom Avanto, Siemens Healthcare, Erlangen, Germany) or a 3 Tesla scanner (Magnetom Skyra, Siemens Healthcare, Erlangen, Germany). The images were archived for evaluation in the hospital’s own picture archiving and communication system (PACS, IMPAX EE R20 XVIII SU1 AGFA, Healthcare N.V., Mortsel, Belgium).

The scans were evaluated independently using the classification systems for forensic age determination by Ottow et al. and Vieth et al. T1-weighted turbo spin echo sequences (TSE) were used for the classification based on Ottow et al. For the evaluation according to the classification based on Vieth et al., in addition to the T1 TSE, T2-weighted sequences with fat suppression in the form of a T2 Turbo Inversion Recovery Magnitude (TIRM) were also analyzed. The sequence parameters for acquiring the MR images are shown in [Table 2].

Table 2 Sequence parameters for the acquisition of MRI scans. TSE: Turbo Spin Echo, TIRM: Turbo Inversion Recovery Magnitude.

	T1 TSE	T2 TIRM
Matrix	484 × 235	384 × 2016
Voxel size	0.4 × 0.4 × 3.0mm	0.3 × 0.4 × 3.0mm
Field of view (FOV)	180mm	200mm
Slice orientation	coronary	coronary
Slice thickness	3mm	3mm
Repetition time (TR)	514ms	5320ms
Echo time (TE)	10ms	42ms
Flip angle	163°	180°
Inversion time		202ms
Acquisition time	2:08min	1:58min

Classifications

Ottow et al. used a classification consisting of five main stages with three sub-stages each in stages 2 and 3 [26] [27]. T1-weighted images were used to assess the ossification of the epiphyseal plate and, if present, the epiphyseal scar. The description of the different ossification stages for this classification is shown in [Table 3]. [Fig. 1] shows sample images used in the present study to assess the ossification of the epiphyseal plates of the upper ankle joint based on the classification of Ottow et al. Due to the age range of the patients included in the study, the classification begins at stage 2c.

Table 3 Ossification stages based on Schmeling et al. and additional substages based on Kellinghaus et al.

Stage	Description
I	The epiphysis has not yet ossified.
II	The epiphysis is partially ossified. The epiphyseal plate is not ossified.
IIa	The longitudinal dimension of the ossified portion of the epiphysis is one third or less compared to the width of the metaphyseal end.
IIb	The longitudinal dimension of the ossified portion of the epiphysis is more than one third to a maximum of two thirds compared to the width of the metaphyseal end.
IIc	The longitudinal dimension of the ossified portion of the epiphysis is more than two thirds compared to the width of the metaphyseal end.
III	The epiphysis is ossified. The epiphyseal plate is partially ossified.
IIIa	The ossification of the epiphyseal plate is one third or less of the initial length.
IIIb	The ossification of the epiphyseal plate is more than one third to a maximum of two thirds of the initial length.
IIIc	The ossification of the epiphyseal plate is more than two thirds of the initial length.
IV	The epiphyseal plate is completely ossified. A joint scar is visible.
V	The epiphyseal plate is completely ossified. A joint scar is no longer visible.

Fig. 1 Ottow et al. Stages. T1 Turbo Spin Echo Sequences (TSE) from the upper ankle joints from the present study. Stage 2c: 12 years female, Stage 3a: 14 years female, Stage 3b: 15 years female, Stage 3c: 15 years male, Stage 4: 19 years male, Stage 5: 25 years female.

Vieth et al. used a six-stage classification system that analyzed T1-weighted images and additionally fat-saturated T2 weighted images [22]. The allocation was made based on the following assessment criteria. Stage 2: Continuous band-like representation of the epiphyseal plate. Stage 3: Discontinuous band-like morphology of the epiphyseal plate. Stage 4: Discontinuous linear joint scar. Stage 5: Joint scar as a continuous line with additional hyper-intense signal in T2. Stage 6: Also linear joint scar in T1 but with missing signal in T2. Stages 5 and 6 can therefore only be distinguished from each other using the T2 sequence. [Fig. 2] shows sample images used in the present study to assess the ossification of the epiphyseal plates of the upper ankle joint based on the classification of Vieth et al.

Fig. 2 Vieth et al. Stages. T1 Turbo Spin Echo Sequences (TSE) from the upper ankle joints from the present study. Stage 2: 12 years female, Stage 3: 14 years female, Stage 4: 15 years male, Stage 5: 19 years male, Stage 6: 25 years female with additional fat saturated T2 Turbo-Inversion Recovery Magnitude (TIRM) sequence.

Before the evaluation, three examiners (A, B, and C) underwent training for the two classification systems using 50 MRI scans of knee joints. Examiners A and B each had one year of experience, and examiner C had 10 years of experience in musculoskeletal radiology. At the end of the training phase, the results were discussed to reach a consensus.

The MR data sets for the study group were analyzed independently by examiner A based on the two classifications. The two classification systems were applied separately to the distal tibia and fibula. The evaluation was carried out after pseudonymization and blinding for chronological age. To determine inter-observer reliability, 74 patients (22% of the total sample) with an even age distribution across the entire study sample were assessed by all three examiners (A, B and C) based on both classification systems. To assess intra-observer reliability, the 74 MRI data sets from examiner A were re-evaluated after three months.

Statistics

The statistical analysis was performed using IBM SPSS Statistics 27 statistical software. Minimum, maximum, mean ± standard deviation, and median with lower and upper quartiles were defined. The ossification stages were presented in relation to the age of the patients in boxplot diagrams. Gender differences were analyzed using the Mann-Whitney U test. The inter-observer and intra-observer reliabilities were calculated using Cohen’s kappa and the Altmann scoring system was used to interpret the kappa values.

Results

Of the initial 447 MRI scans, a total of 333 data sets remained after applying the described criteria (160 female, 173 male). For the evaluation based on Vieth et al., only 324 patients were included due to a lack of suitable fat-saturated T2-weighted images ([Fig. 3]). The patients were divided into four age groups according to their chronological age (<14 years; 14–17 years; 18–20 years; >21 years). [Table 4] shows the frequency of age groups for the 333 patients in both sexes. A total of 167 MRI scans of the left ankle and 166 scans of the right ankle were evaluated.

Table 4 Frequency of age groups among male and female patients.

Male				Female
Age	Frequency	Percent	Cumulative percentages	Age	Frequency	Percent	Cumulative percentages
<14	37	21.4	21.4	<14	37	23.1	23.1
14–17	40	23.1	44.5	14–17	56	35.0	58.1
18–20	20	11.6	56.1	18–20	12	7.5	65.6
>/=21	76	43.9	100.0	>/=21	55	34.4	100.0
Total	173	100.0		Total	160	100.0

Fig. 3 Flowchart for determining the patient group.

Evaluation based on Ottow et al.

Age limit 14 years old

When evaluating the tibial epiphysis based on the classification by Ottow et al., the patients classified in stage 2c were younger than 14 years. All patients who were classified as stage 3b or higher when evaluating the tibial epiphysis and fibular epiphysis were older than 14 years ([Fig. 4]). When evaluating the fibular epiphysis for comparison with the tibial epiphysis, not all patients in stage 2c were younger than 14 years of age. The other results that fall below or exceed the relevant age limits were otherwise very similar when evaluating the fibula ([Fig. 4]). The specificity for the age limit of 14 years was 100% for both epiphyseal plates from stage 3b onwards.

Fig. 4 Boxplot: Ossification stages of the distal tibia and fibula in relation to the age of male patients according to the classification of Ottow et al.

The patients who were classified in stage 2c were also all younger than 14 years ([Fig. 5]). In addition, when evaluating the female fibular epiphysis, not all patients in stage 2c were younger than 14 years, compared to the tibial epiphysis. After applying the minimum age concept, the completion of 14 years of age could not be determined for both epiphyseal plates ([Fig. 5]). The specificity for reaching the age of 14 years from stage 4 or higher was 98.18% for the fibular epiphysis and 94.59% for the tibial epiphysis.

Fig. 5 Boxplot: Ossification stages of the distal tibia and fibula in relation to the age of female patients according to the classification of Ottow et al.

Age limit 18 years old

The patients who were classified in stages 2c and 3a, based on the tibial epiphysis, were younger than 18 years ([Fig. 4]). With the exception of one outlier, all patients in stage 3b were under 18 years of age. The patients who were classified in stages 2c to 3b were all under 18 years of age. In stage 5, all patients – with one exception – were older than 18 years ([Fig. 5]). In both male and female patients, the age of majority could not be determined using the minimum age concept for both epiphyseal plates. When evaluating the fibular epiphysis, no significant differences were found between the sexes with regard to those over or under 18 years of age. The specificity in male patients from stage 5 onwards was 98.70% at the tibial epiphysis and 97.70% at the fibular epiphysis. In female patients, the specificity was 98.92% at the tibial epiphysis and 94.62% at the fibular epiphysis.

Age limit 21 years old

At the tibial epiphysis, with the exception of one outlier in stage 3b, all patients classified in stages 2c to 3c were younger than 21 years of age ([Fig. 4]). Among the female patients, with the exception of one outlier in stage 3c, all patients classified in stages 2c to 3c were also under 21 years of age ([Fig. 5]). When evaluating the fibular epiphysis, no significant differences were found between the sexes with regard to those over or under 21 years of age. The specificity for the 21st year of life was as follows. Male: tibial epiphysis 96.90% and fibular epiphysis 89.69%; female: tibial epiphysis 99.05% and fibular epiphysis 89.52%.

The frequencies of the respective ossification stages for the tibia and fibula, as well as minimum and maximum chronological age per stage, are shown in [Table 5] and [Table 6] for both sexes. Supplemental tables (Tables A1 and A2) show the frequency of ossification stages in the age groups for tibia and fibula.

Table 5 Descriptive statistic of the ossification stages of the tibia and fibula based on Ottow et al., male.

Tibia and fibula stage/age (in years), male
		No.	Min.	Max.	Median			No.	Min.	Max.	Median
Tibia stage	2c	23	10.38	13.61	11.76	Fibula stage	2c	28	10.61	14.58	12.17
	3a	27	12.12	16.74	13.83		3a	25	10.38	16.97	14.14
	3b	8	14.94	23.94	15.96		3b	4	14.94	17.18	15.83
	3c	11	14.53	18.60	16.38		3c	14	14.53	23.94	16.89
	4	89	14.88	28.93	23.33		4	42	14.88	28.44	22.22
	5	15	17.52	28.73	24.53		5	60	16.25	28.93	24.73

Table 6 Descriptive statistic of the ossification stages of the tibia and fibula based on Ottow et al., female.

Tibia and fibula stage/age (in years), female
		No.	Min.	Max.	Median			No.	Min.	Max.	Median
Tibia stage	2c	12	10.19	12.69	11.73	Fibula stage	2c	21	10.19	14.16	12.03
	3a	18	10.58	14.89	13.09		3a	14	12.16	16.02	14.10
	3b	8	12.27	16.02	14.19		3b	11	12.27	15.77	13.23
	3c	21	12.73	22.09	14.62		3c	21	13.04	18.36	15.39
	4	83	13.65	28.73	18.93		4	46	14.44	28.57	18.15
	5	18	17.41	28.92	25.88		5	47	13.74	28.92	24.11

Using the Mann-Whitney U test, significant differences were found for the respective stages between the sexes. There were significant differences in the tibia between stages 3a and 4 (3a: p=0.037; 3b: p=0.007; 3c: p=0.006; 4: p=<0.001), in the fibula in stages 3b to 4 (3b: p=0.01; 3c: p=0.034; 4: p=0.023). Overall, the differences can be summarized as follows: in percentage terms, male patients tend to be older than female patients in the respective stages.

Evaluation based on Vieth et al.

Age limit 14 years old

When evaluating the tibial epiphysis of the patients based on the classification by Vieth et al., the patients who were classified in stage 2 were under 14 years of age ([Fig. 6]). All patients, both in the tibial and fibular epiphysis evaluation, who were classified as stage 4 or higher, were older than 14 years. When evaluating the fibular epiphysis, compared to the tibia, not all patients in stage 2 were younger than 14 years of age ([Fig. 6]).

Fig. 6 Boxplot: Ossification stages of the distal tibia and fibula in relation to the age of the male patients according to the classification of Vieth et al.

The patients who were classified as stage 2 based on the tibial epiphysis were not yet 14 years old ([Fig. 7]). In comparison to the tibia, not all patients in the fibula evaluations were younger than 14 years. All patients with both epiphyseal plates classified in stages 5 and 6 were older than 14 years. The remaining classifications in the respective stages showed no significant difference with regard to the age limit of 14 years ([Fig. 7]). The specificity for reaching 14 years of age from stage 4 or higher was 100% in both sexes.

Fig. 7 Boxplot: Ossification stages of the distal tibia and fibula in relation to the age of the female patients according to the classification of Vieth et al.

Age limit 18 years old

When assessing the tibial epiphysis, all patients classified in stages 2 and 3 were under the age of 18. The patients whose tibial epiphysis was assigned to stage 6 were all of legal age ([Fig. 6]), but this was not the case for all patients when examining the fibular epiphysis, where 95.4% of the patients in stage 6 were older than 18 years.

All patients who were classified as stage 2 to 3 based on the tibial epiphysis were not yet of legal age. There was no stage in which all of the patients were older than 18 years ([Fig. 7]). 90.3% of the patients classified as stage 6 based on the tibial epiphysis were of adult age, and 84.2% of the patients classified as stage 6 based on the fibular epiphysis. The specificity for the tibial epiphysis of male patients at age 18 was 100% and for the fibular epiphysis it was 96.10%. In female patients, the specificity was 96.67% for the tibial epiphysis and 90.00% for the fibular epiphysis.

Age limit 21 years old

When assessing the tibial epiphysis, all patients in stages 2 to 4 were younger than 21 years ([Fig. 6]). However, after applying the minimum age concept, the completion of the 21st year could not be determined for any stage; for the tibial epiphysis, 92.9% were in stage 6 21 years or older, and for the fibular epiphysis, 83.1%. All patients who were classified in stages 2 to 4 based on the tibial epiphysis were not yet of legal age and were therefore younger than 21 years ([Fig. 7]). In the female patients, the completion of the 21st year of life could not be determined based on the minimum age concept; in the case of the tibial epiphysis, 77.4% were ≥21 years in stage 6, and for the fibular epiphysis, 70.2%. The specificity for the 21st year of life was as follows. Male: tibial epiphysis 97.92% and fibular epiphysis 88.54%; female: tibial epiphysis 93.07% and fibular epiphysis 83.17%.

The frequencies of the respective ossification stages for the tibia and fibula, as well as minimum and maximum chronological age per stage, are shown in [Table 7] and [Table 8]. Supplemental tables (Tables A3 and A4) show the frequency of ossification stages in the age groups for tibia and fibula.

Table 7 Descriptive statistic of the ossification stages of the tibia and fibula based on Vieth et al., male.

Tibia and fibula stage/age (in years), male
		No.	Min.	Max.	Median			No.	Min.	Max.	Median
Tibia stage	2	9	10.38	13.61	11.88	Fibula stage	2	21	10.61	14.58	12.21
	3	48	10.61	17.18	13.68		3	36	10.38	17.18	14.08
	4	8	14.53	18.08	16.66		4	8	14.53	23.43	16.82
	5	78	14.88	28.75	22.67		5	41	14.88	28.75	20.55
	6	28	18.37	28.93	25.54		6	65	16.25	28.93	24.67

Table 8 Descriptive statistic of the ossification stages of the tibia and fibula based on Vieth et al., female.

Tibia and fibula stage/age (in years), female
		No.	Min.	Max.	Median			No.	Min.	Max.	Median
Tibia stage	2	7	10.19	12.06	11.69	Fibula stage	2	14	10.19	14.16	11.73
	3	29	10.36	16.02	12.76		3	26	11.29	16.02	13.16
	4	17	12.73	15.77	14.09		4	16	12.73	16.62	14.59
	5	69	14.44	28.73	18.11		5	40	14.44	27.48	17.26
	6	31	15.73	28.92	25.06		6	57	15.31	28.92	23.84

Using the Mann-Whitney U test, significant differences were found for the respective stages between the sexes. For the tibia, there were significant differences in stage 4 (p=0.002) and in stage 5 (p=0.001). For the fibula, there were significant differences in stage 2 (p=0.037), stage 4 (p=0.013), and stage 5 (p=0.013). Here, too, the differences can be summarized as follows: female patients were younger the male patients in the respective stages.

Intra- and inter-observer reliability

The value for intra-observer reliability in the classification based on Ottow et al. was very good with values of 0.980 for the tibia and 0.927 for the fibula. The inter-observer reliability in the classification based on Ottow et al. between the examiners was between 0.717 and 0.981, which corresponds to a rating of good to very good according to the verbal assessment based on Altmann.

The value for intra-observer reliability in the classification based on Vieth et al. was very good for both the tibia (0.872) and the fibula (0.903). The value for the inter-observer reliability in the classification based on Vieth et al. between the examiners was between 0.885 and 0.942, which corresponds to a rating of very good according to the verbal assessment based on Altmann.

Discussion

The results show that by assessing the distal tibia with both classifications, the age of 14 years can be determined for both sexes and the age of 18 years in male patients with the classification based on Vieth et al. The age of 18 years could be determined using both classification systems for both sexes and bones (tibia and fibula). Applying the classification of Vieth et al., we were able to use the tibia and the fibula to determine when male patients had reached the age of 21 years.

Based on 180 MRI data sets, St. Martin et al. developed a classification system with three stages to assess the ossification of the distal tibia and the calcaneus [19]. In another study, St. Martin et al. were able to correctly classify 97.4% of male and 93.4% of female participants who were 18 years or older in their cohort of 160 people [20]. The 3-stage system was expanded to six stages by Lu et al. and applied to 590 subjects 8 to 25 years of age [17]. The 6-stage classification was used in another study by Yavuc et al. [28]. The authors concluded that this classification can be used as a complementary method for determining the age limit of 18 years. In addition to the three-stage classification by St. Martin et al. for the ankle joint and the classification extended to six stages by Lu et al., Gurses et al. applied the classification of Vieth et al. for MR data sets related to the ankle joint [18] [19] [20] [22]. In their cohort, this classification made it possible to determine the age of 18. In addition to the age of majority, the age limits of 14 or 21 were not considered separately in these studies.

Age limit 14 years old

Based on the minimum age concept, it was possible in our study to determine the completion of the 14^th year for the tibia and the fibula in male patients using the classification by Ottow et al. with the classification in stage 3b. In female patients, this was also possible using the tibia with classification in stage 5. For the fibula in female patients and the evaluation based on Ottow et al., this was not possible in our study; there was an outlier in stage 5 at just under 14 years of age. In their study, Ottow et al., on the other hand, were able to determine the completion of the 14^th year in female patients with stage 3b for the femur and stage 4 for the proximal tibia when evaluating MR images of the knee joint. In our study group, completion of the 14^th year was evident in male patients from stage 4 onwards at both the distal tibia and the distal fibula. For female patients, completion of the 14^th year of life was evident from stage 5 onwards. The specificity regarding the minimum age concept was thus 100% for stage 4 in male patients and 100% for stage 5 in female patients. Using the classification based on Ottow et al., a specificity of 100% was achieved only for stage 3b in male patients. This age limit is not explicitly addressed in the study by Vieth et al. However, looking at the data, the results of Vieth et al. with regard to the completion of the 14^th year of life are similar to those in the present study. One difference is that in Vieth et al., the age of 14 years in female patients was not defined by classification in stage 4 for the femur and tibia, but by classification in stage 5 for the proximal tibia. Gurses et al. also applied the classification by Vieth et al. to bones of the ankle joint [18]. According to Gurses et al., the 14^th year of life can be defined retrospectively in both sexes by classification in stage 5 for the tibia and for the calcaneus classification in stage 5 in male patients and stage 6 for female patients. One potential explanation could be differences in ethnic background between the original publications, the present study, and the study group in Gurses et al.

Age limit 18 years old

Similar to the original publication, our study also could not determine the completion of the 18^th year using the evaluation based on Ottow et al. In the evaluation based on Vieth et al., we could only determine the age of majority in our study for male patients using the tibia. In contrast, in the study by Vieth et al., the age of majority could be determined for the femur and tibia in male patients and by analyzing the femur in the female patients. In contrast to Gurses et al., in which all patients in stages 2–4 were younger than 18 years of age, in our case this was only true for the female patients. In our study, the age of 18 years could also be determined for the tibia of the male patients. In Gurses et al., the age of majority could not be determined. Considering the specificity with regard to the minimum age concept for the age limit of 18 years, the values in our study for the classification based on Ottow et al. were between 94.62% and 98.92%, and for the classification by Vieth et al. between 90.00% and 100%.

Age limit 21 years old

In order to also evaluate the age limit of 21 years, the maximum age for study inclusion was defined as 28 years in the present study compared to the original publication. In the classification based on Vieth et al., all female and male patients in stages 2 to 4 for both the tibia and the fibula were younger than 21 years. In the classification based on Ottow et al., all patients in stages 2c to 3c were younger than 21 years, with one exception. Due to the age structure of the study group, the limit of 21 years of age was not considered separately in the original publications. Due to the inclusion criteria regarding chronological age, the results of the present study also allow conclusions to be drawn about the limit of 21 years of age for both classifications.

Overall, the average age was shown to increase continuously in stages 2c to 5 in Ottow et al. and stages 2 to 6 in Vieth et al., which suggests that the stages actually reflect the progressive ossification of the epiphyseal plate. Only in the case of the fibula for female patients in the Ottow evaluation is the average age in stage 3b lower than that in stage 3a. This could imply that the substages 3a to 3c of Ottow et al. differ too little from each other, which is supported by the fact that the average age of the three stages for all bones is very similar in both sexes and larger jumps are found only from stage 2c to stage 3a, and particularly from stage 3c to stage 4.

The TIRM sequence used in the present study for the classification by Vieth et al. was chosen in order to use the same sequences as in the original publication. More common sequences in MRI scans of the ankle – such as fat-saturated proton-weighted sequences – have already been used successfully in recent studies [25] [29].

In contrast to the studies by Ottow et al. and Vieth et al., the present study retrospectively evaluated data sets that were not explicitly created for age estimation. It was shown that these two classification systems can also be applied with comparable results, using MRI scans of the distal tibia and fibula that were first created in answer to other medical questions. The ability to apply the classifications to another joint without any issues is supported by the fact that the classifications are based on the investigation of general milestones of the ossification process for epiphyseal plates and not on features of specific bones.

The scan of the ankle joint as a peripheral joint could reduce the burden during forensic age determination, since it is no longer necessary to position the person to be examined fully in the MR scanner. The use of portable MR scanners to examine the ankle joint in contrast to other, more centrally located body regions for age determination, such as the knee joint or the clavicle, would be another option. A study on the use of the classification by Vieth et al. in low-field MRI (0.31 T) already exists [25].

Outlook

The use of MRI as a radiation-free alternative to forensic age diagnostics has been investigated in numerous studies [30]. Due to the currently insufficient reference data and possibly the sometimes long scanning times, potential contraindications, higher costs, and the increased requirement for compliance, there is currently no recommendation from the AGFAD [2]. The classifications used in the present study can be considered a promising approach. Current validation studies to also improve applicability, particularly in the low-field range, show promising results [25] [31]. To improve the reference data, additional studies are needed with sufficiently large groups and covering a wide range of ages.

Limitations

With regard to limitations, the first thing to mention is the lack of recording of systemic diseases and medications that could affect the ossification process. These include growth disorders, hormone therapy, steroids, and completed chemotherapy, which can influence the ossification of the epiphyseal plate. In cases of pathologies that hindered assessment of the epiphyseal plates, those scans were excluded.

Limitations of our study include the retrospective study design and the unequal age distribution in the different groups.

A further limitation is that not all data sets were evaluated by all examiners. The different experience levels of the examiners should also be mentioned, although for bone age determination using the Greulich/Pyle method, the experience of the examiners in direct comparison did not show any relevant influence on the accuracy of the evaluation [32].

Furthermore, in the present study, the nationality of the patients and their socioeconomic status could not be identified, retrospectively. This made it impossible to take into account possible ethnic differences in bone growth.

Conclusions

In the present study, it was possible to apply the classifications for age determination based on Ottow et al. and Vieth et al., which were developed primarily using MRI of the knee joint, to the ankle joint as well and to determine completion of the 14^th year for both sexes and completion of the 18^th year for male patients using the classification based on Vieth et al. Furthermore, the age of majority could be determined using both classification systems for both sexes and bones (tibia and fibula). Conclusions about reaching the age of 21 were also possible. The present study shows that MR images of the ankle joint represent a possible radiation-free alternative to X-rays of the hand or CT images of the clavicle, and they can potentially be used as an additional assessment criterion in forensic age diagnostics.

Clinical Relevance of Study

• MRI of the ankle is a radiation-free alternative for forensic age diagnostics.

• Classifications that were primarily developed using MR images of the knee joint can also be applied to the ankle joint.

• The assessment of the ossification of the epiphyseal plates of the ankle joint can help to determine when legal age limits have been reached.

Conflict of Interest

The authors declare that they have no conflict of interest.

• References

• 1 Lockemann U, Fuhrmann A, Püschel K. et al. Arbeitsgemeinschaft für Forensische Altersdiagnostik der Deutschen Gesellschaft für Rechtsmedizin. Rechtsmedizin 2004; 14: 123-126
• 2 Wittschieber D, Hahnemann ML, Mentzel HJ. Forensic Diagnostics of the Skeletal Age in the Living – Backgrounds and Methodology. Fortschr Röntgenstr 2024; 196: 254-261
• 3 Befurt L, Kirchhoff G, Rudolf E. et al. Juristische Aspekte der forensischen Altersdiagnostik auf der Grundlage des § 42f SGB VIII. Rechtsmedizin 2020; 30: 241-248
• 4 Schmeling A, Dettmeyer R, Rudolf E. et al. Forensic Age Estimation. Dtsch Arztebl Int 2016; 113: 44-50
• 5 Schmeling A, Püschel K. Forensische Altersdiagnostik. Rechtsmedizin 2010; 20: 457-458
• 6 Schmeling A, Püschel K. Forensische Altersdiagnostik. Rechtsmedizin 2011; 21: 5-6
• 7 Schmeling A, Püschel K. Forensische Altersdiagnostik III. Rechtsmedizin 2014; 24: 457-458
• 8 Schmeling A, Schulz R, Danner B. et al. The impact of economic progress and modernization in medicine on the ossification of hand and wrist. International Journal of Legal Medicine 2006; 120: 121-126
• 9 Olze A, Bilang D, Schmidt S. et al. Validation of common classification systems for assessing the mineralization of third molars. International journal of legal medicine 2005; 119: 22-26
• 10 Liversidge HM. The assessment and interpretation of Demirjian, Goldstein and Tanner’s dental maturity. Annals of human biology 2012; 39: 412-431
• 11 Schmeling A, Schulz R, Reisinger W. et al. Studies on the time frame for ossification of the medial clavicular epiphyseal cartilage in conventional radiography. International Journal of Legal Medicine 2004; 118: 5-8
• 12 Focardi M, Pinchi V, De Luca F. et al. Age estimation for forensic purposes in Italy: Ethical issues. International Journal of Legal Medicine 2014; 128: 515-522
• 13 Urschler M, Krauskopf A, Widek T. et al. Applicability of Greulich–Pyle and Tanner–Whitehouse grading methods to MRI when assessing hand bone age in forensic age estimation: A pilot study. Forensic Science International 2016; 266: 281-288
• 14 Hojreh A, Gamper J, Schmook MT. et al. Hand MRI and the Greulich-Pyle atlas in skeletal age estimation in adolescents. Skeletal Radiology 2018; 47: 963-971
• 15 Diete V, Wabitsch M, Denzer C. et al. Applicability of magnetic resonance imaging for bone age estimation in the context of medical issues. Fortschr Röntgenstr 2021; 193: 692-700
• 16 Ekizoglu O, Hocaoglu E, Can IO. et al. Magnetic resonance imaging of distal tibia and calcaneus for forensic age estimation in living individuals. International Journal of Legal Medicine 2015; 129: 825-831
• 17 Lu T, Shi L, Zhan M-j. et al. Age estimation based on magnetic resonance imaging of the ankle joint in a modern Chinese Han population. International Journal of Legal Medicine 2020; 134: 1843-1852
• 18 Gurses MS, Has B, Altinsoy HB. et al. Evaluation of distal tibial epiphysis and calcaneal epiphysis according to the Vieth method in 3.0 T magnetic resonance images: A pilot study. International Journal of Legal Medicine 2023; 137: 1181-1191
• 19 Saint-Martin P, Rérolle C, Dedouit F. et al. Age estimation by magnetic resonance imaging of the distal tibial epiphysis and the calcaneum. International Journal of Legal Medicine 2013; 127: 1023-1030
• 20 Saint-Martin P, Rérolle C, Dedouit F. et al. Evaluation of an automatic method for forensic age estimation by magnetic resonance imaging of the distal tibial epiphysis—a preliminary study focusing on the 18-year threshold. International Journal of Legal Medicine 2014; 128: 675-683
• 21 Ottow C, Schulz R, Pfeiffer H. et al. Forensic age estimation by magnetic resonance imaging of the knee: The definite relevance in bony fusion of the distal femoral- and the proximal tibial epiphyses using closest-to-bone T1 TSE sequence. Eur Radiol 2017; 27: 5041-5048
• 22 Vieth V, Schulz R, Heindel W. et al. Forensic age assessment by 3.0T MRI of the knee: Proposal of a new MRI classification of ossification stages. European Radiology 2018; 28: 3255-3262
• 23 Dodson RM, Firoozmand A, Hyder O. et al. Impact of sarcopenia on outcomes following intra-arterial therapy of hepatic malignancies. J Gastrointest Surg 2013; 17: 2123-2132
• 24 Terada Y, Kono S, Tamada D. et al. Skeletal age assessment in children using an open compact MRI system. Magn Reson Med 2013; 69: 1697-1702
• 25 Ottow C, Schmidt S, Schulz R. et al. Forensische Altersdiagnostik mittels Niederfeld-Magnetresonanztomographie. Rechtsmedizin 2023; 33: 96-104
• 26 Kellinghaus M, Schulz R, Vieth V. et al. Enhanced possibilities to make statements on the ossification status of the medial clavicular epiphysis using an amplified staging scheme in evaluating thin-slice CT scans. International Journal of Legal Medicine 2010; 124: 321-325
• 27 Ottow C, Schulz R, Pfeiffer H. et al. Forensic age estimation by magnetic resonance imaging of the knee: The definite relevance in bony fusion of the distal femoral- and the proximal tibial epiphyses using closest-to-bone T1 TSE sequence. European Radiology 2017; 27: 5041-5048
• 28 Yavuz TK, Hilal A, Kaya O. et al. Forensic age estimation with ankle MRI: Evaluating distal tibial and calcaneal epiphyseal fusion. Forensic Sci Int 2023; 352: 111832
• 29 Chitavishvili N, Papageorgiou I, Malich A. et al. The distal femoral epiphysis in forensic age diagnostics: Studies on the evaluation of the ossification process by means of T1- and PD/T2-weighted magnetic resonance imaging. Int J Legal Med 2023; 137: 427-435
• 30 De Tobel J, Bauwens J, Parmentier GIL. et al. Magnetic resonance imaging for forensic age estimation in living children and young adults: A systematic review. Pediatr Radiol 2020; 50: 1691-1708
• 31 Wittschieber D, Chitavishvili N, Papageorgiou I. et al. Magnetic resonance imaging of the proximal tibial epiphysis is suitable for statements as to the question of majority: A validation study in forensic age diagnostics. Int J Legal Med 2022; 136: 777-784
• 32 Lynnerup N, Belard E, Buch-Olsen K. et al. Intra- and interobserver error of the Greulich-Pyle method as used on a Danish forensic sample. Forensic Sci Int 2008; 179: 242.e241-e246

Publikationsverlauf

Eingereicht: 19. April 2024

Angenommen nach Revision: 29. Juli 2024

Artikel online veröffentlicht:
05. September 2024