Comparing Traditional Age Estimation at the Defense POW/MIA Accounting Agency to Age Estimation Using Random Forest Regression
Abstract
:1. Introduction
2. Materials and Methods
3. Results
4. Discussion
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Blau, S.; Ubelaker, D.H. (Eds.) Handbook of Forensic Anthropology and Archaeology; Routledge: New York, NY, USA, 2016. [Google Scholar]
- Christensen, A.M.; Passalacqua, N.V.; Bartelink, E.J. Forensic Anthropology: Current Methods and Practice; Elsevier Inc.: San Diego, CA, USA, 2014. [Google Scholar]
- Latham, K.E.; Finnegan, M. (Eds.) Age Estimation of the Human Skeleton; Charles C. Thomas: Springfield, IL, USA, 2010. [Google Scholar]
- Rogers, T.L. Skeletal age estimation. In Handbook of Forensic Anthropology and Archaeology; Blau, S., Ubelaker, D.H., Eds.; Routledge: New York, NY, USA, 2016; pp. 273–292. [Google Scholar]
- Todd, T.W. Age changes in the pubic bone I. The male white pubis. Am. J. Phys. Anthropol. 1920, 3, 285–334. [Google Scholar] [CrossRef]
- Todd, T.W. Age changes in the pubic bone. Am. J. Phys. Anthropol. 1921, 4, 1–77. [Google Scholar] [CrossRef]
- Katz, D.; Suchey, J.M. Age determination of the male os pubis. Am. J. Phys. Anthropol. 1986, 69, 427–435. [Google Scholar] [CrossRef]
- Berg, G.E. Pubic bone age estimation in adult women. J. Forensic Sci. 2008, 53, 569–577. [Google Scholar] [CrossRef]
- Hartnett, K.M. Analysis of age at death estimation using data from a new, modern autopsy sample—Part I: Pubic bone. J. Forensic Sci. 2010, 55, 1145–1151. [Google Scholar] [CrossRef]
- Lovejoy, C.O.; Meindl, R.S.; Pryzbeck, T.R.; Mensforth, R.P. Chronological metamorphosis of the auricular surface of the ilium: A new method for the determination of adult skeletal age at death. Am. J. Phys. Anthropol. 1985, 68, 15–28. [Google Scholar] [CrossRef]
- Buckberry, J.L.; Chamberlain, A.T. Age estimation from the auricular surface of the ilium: A revised method. Am. J. Phys. Anthropol. 2002, 119, 231–239. [Google Scholar] [CrossRef]
- Osborne, D.L.; Simmons, T.L.; Nawrocki, S.P. Reconsidering the auricular surface as an indicator of age at death. J. Forensic Sci. 2004, 49, 905–911. [Google Scholar] [CrossRef]
- Boldsen, J.L.; Milner, G.R.; Konigsberg, L.W.; Wood, J.W. Transition analysis: A new method for estimating age from skeletons. In Paleodemography: Age Distributions from Skeletal Samples; Hoppa, R.D., Vaupel, J.W., Eds.; Cambridge University Press: Cambridge, UK, 2002; pp. 73–106. [Google Scholar]
- Getz, S.M. The use of transition analysis in skeletal age estimation. Wiley Interdiscip. Rev. Forensic Sci. 2020, 2, c1378. [Google Scholar] [CrossRef]
- Garvin, H.M.; Passalacqua, N.V. Current practices by forensic anthropologists in adult skeletal age estimation. J. Forensic Sci. 2013, 57, 427–433. [Google Scholar] [CrossRef] [PubMed]
- Ubelaker, D.H.; Khosrowshahi, H. Estimation of age in forensic anthropology: Historical perspective and recent methodological advances. Forensic Sci. Res. 2019, 4, 1–9. [Google Scholar] [CrossRef] [PubMed]
- Katz, D.; Suchey, J. Race differences in pubic symphyseal aging patterns in the male. Am. J. Phys. Anthropol. 1989, 80, 167–172. [Google Scholar] [CrossRef]
- Hoppa, R.D. Population variation in osteological aging criteria: An example from the pubic symphysis. Am. J. Phys. Anthropol. 2000, 111, 185–191. [Google Scholar] [CrossRef]
- Langley-Shirley, N.; Jantz, R.L. A Bayesian approach to age estimation in modern Americans from the clavicle. J. Forensic Sci. 2010, 55, 571–583. [Google Scholar] [CrossRef]
- Suchey, J.M. Problems in the aging of females using the os pubis. Am. J. Phys. Anthropol. 1979, 51, 467–470. [Google Scholar] [CrossRef]
- Suchey, J.M.; Brooks, S.T.; Katz, D. Instructions for use of the Suchey-Brooks system for age determination of the female os pubis. In Instructional Materials Accompanying Female Pubic Symphysial Models of the Suchey-Brooks System; France Casting: Fort Collins, CO, USA, 1988. [Google Scholar]
- Bello, S.M.; Thomann, A.; Signoli, M.; Dutour, O.; Andrews, P. Age and sex bias in the reconstruction of past population structures. Am. J. Phys. Anthropol. 2006, 129, 24–38. [Google Scholar] [CrossRef] [PubMed]
- Scheuer, L.; Black, S. Developmental Juvenile Osteology, 1st ed.; Academic Press: New York, NY, USA, 2000. [Google Scholar]
- Schaefer, M.C.; Black, S.M. Comparison of ages of epiphyseal union in north American and Bosnian skeletal material. J. Forensic Sci. 2005, 50, 777–784. [Google Scholar] [CrossRef] [PubMed]
- McKern, T.W.; Stewart, T.D. Skeletal Age Changes in Young American Males; Technical Report EP-45; Quartermaster Research and Development Command: Natick, MA, USA, 1957; Available online: https://apps.dtic.mil/sti/citations/AD0147240 (accessed on 2 March 2022).
- Wilson, E.K. A125: A reanalysis of Korean War anthropological records to support the resolution of cold cases. In Proceedings of the American Academy of Forensic Sciences, Las Vegas, NV, USA, 22–27 February 2019; Peat, M.A., Ed.; American Academy of Forensic Sciences: Colorado Springs, CO, USA, 2019; pp. 189–196. [Google Scholar]
- Taylor, R.J.; Scott, A.L.; Koehl, A.J.; Trask, W.R.; Maijanen, H. The Tarawa Project Part I: A multidisciplinary approach to resolve commingled human remains from the Battle of Tarawa. Forensic Anthropol. 2019, 2, 87–95. [Google Scholar] [CrossRef]
- Defense POW/MIA Accounting Agency. Battle of Tarawa. Available online: https://dpaa-mil.sites.crmforce.mil/dpaaFamWebInTarawa (accessed on 20 February 2023).
- Breiman, L. Random forest. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef]
- R Core Team. R. A Language and Environment for Statistical Computer, R Foundation for Statistical Computing: Vienna, Austria, 2021. Available online: https://www.R-project.org/(accessed on 30 January 2023).
- Kuhn, M. Building Predictive Models in R Using the caret Package. J. Stat. Softw. 2008, 28, 1–26. [Google Scholar] [CrossRef]
- Hoppa, R.D.; Vaupel, J.W. The Rostock Manifesto for paleodemography: The way from stage to age. In Paleodemography: Age Distributions from Skeletal Samples; Hoppa, R.D., Vaupel, J.W., Eds.; Cambridge University Press: Cambridge, UK, 2002; pp. 1–8. [Google Scholar]
- Hoppa, R.D.; Vaupel, J.W. Paleodemography: Age Distributions from Skeletal Samples; Cambridge University Press: Cambridge, UK, 2002. [Google Scholar]
- Konigsberg, L.W. Multivariate cumulative probit for age estimation using ordinal categorical data. Ann. Hum. Biol. 2015, 42, 368–378. [Google Scholar] [CrossRef] [PubMed]
- Usher, B.M. Reference samples: The first step in linking biology and age in the human skeleton. In Paleodemography; Hoppa, R.D., Vaupel, J.W., Eds.; Cambridge University Press: Cambridge, UK, 2018; pp. 29–47. [Google Scholar]
- Navega, D.; Costa, E.; Cunha, E. Adult Skeletal Age-at-Death Estimation through Deep Random Neural Networks: A New Method and Its Computational Analysis. Biology 2022, 11, 532. [Google Scholar] [CrossRef] [PubMed]
Area/Element | Indicator |
---|---|
Clavicle | Medial Epiphysis |
Pubic Symphysis | Dorsal Demiface |
Ventral Demiface | |
Symphyseal Rim | |
Os Coxa | Ischiopubic Ramus |
Iliac Crest | |
Sacrum | S1/S2 |
S2/S3 | |
Scapula | Acromion |
Medial border | |
Lateral border | |
Limbs | Proximal Humerus |
Distal Ulna | |
Distal Radius | |
Distal Femur | |
Proximal Tibia | |
Distal Fibula |
Component | Description |
---|---|
Dorsal Demiface | |
Stage 0 | Dorsal margin absent. |
Stage 1 | A slight margin formation first appears in the middle third of the dorsal border. |
Stage 2 | The dorsal margin extends along entire dorsal border. |
Stage 3 | Filling in of grooves and resorption of ridges to form a beginning plateau in the middle third of the dorsal demiface. |
Stage 4 | The plateau, still exhibiting vestiges of billowing, extends over most of the dorsal demiface. |
Stage 5 | Billowing disappears completely and the surface of the entire demiface becomes flat and slightly granulated in texture. |
Ventral Demiface | |
Stage 0 | Ventral beveling is absent. |
Stage 1 | Ventral beveling is present only at superior extremity of ventral border. |
Stage 2 | Bevel extends inferiorly along ventral border. |
Stage 3 | The ventral rampart begins by means of bony extensions from either or both of the extremities. |
Stage 4 | The rampart is extensive, but gaps are still evident along the earlier ventral border, most evident in the upper two-thirds. |
Stage 5 | The rampart is complete. |
Symphyseal Rim | |
Stage 0 | The symphyseal rim is absent |
Stage 1 | A partial dorsal rim is present, usually at the superior end of the dorsal margin; it is round and smooth in texture and elevated above the symphyseal surface. |
Stage 2 | The dorsal rim is complete and the ventral rim is beginning to form. There is no particular beginning site. |
Stage 3 | The symphyseal rim is complete. The enclosed symphyseal surface is finely grained in texture and irregular or undulating in appearance. |
Stage 4 | The rim begins to break down. The face becomes smooth and flat and the rim is no longer round but sharply defined. There is some evidence of lipping on the ventral edge. |
Stage 5 | Further breakdown of the rim (especially along superior ventral edge) and rarefaction of the symphyseal face. There is also disintegration and erratic ossification along the ventral rim. |
Ind. | Known Age | FAR Analysis * | McKern and Stewart * | Total Sample * | |||||
---|---|---|---|---|---|---|---|---|---|
Age | Interval | RF Est. | 95% PI | Interval | RF Est. | 95% PI | Interval | ||
1 | 38.3 | 23–39 | 17 | 35.1 | 31–38 | 8 | 35.0 | 31–39 | 9 |
2 | 35.4 | 24–39 | 16 | 28.8 | 24–31 | 8 | 28.7 | 25–31 | 7 |
3 | 34.9 | 23–39 | 17 | 29.1 | 24–33 | 10 | 29.2 | 24–33 | 10 |
4 | 33.5 | 23–39 | 17 | 30.5 | 29–32 | 4 | 30.5 | 28–32 | 5 |
5 | 33.1 | 23–39 | 17 | 31.9 | 29–33 | 5 | 31.9 | 29–33 | 5 |
6 | 31.8 | 23–39 | 17 | 31.9 | 29–33 | 5 | 31.8 | 29–33 | 5 |
7 | 28.8 | 23–35 | 13 | 29.1 | 27–31 | 5 | 29.0 | 27–31 | 5 |
8 | 26.8 | 22–28 | 7 | 27.0 | 22–31 | 10 | 26.8 | 22–31 | 10 |
9 | 26.1 | 23–36 | 14 | 33.2 | 28–37 | 10 | 33.5 | 28–38 | 11 |
10 | 24.3 | 23–33 | 11 | 27.6 | 23–35 | 13 | 27.6 | 23–35 | 13 |
11 | 24.0 | 23–39 | 17 | 30.5 | 29–32 | 4 | 30.7 | 29–32 | 4 |
12 | 23.7 | 22–30 | 9 | 26.2 | 22–38 | 17 | 26.3 | 22–39 | 18 |
13 | 23.3 | 20–28 | 9 | 24.2 | 22–25 | 4 | 24.1 | 22–25 | 4 |
14 | 23.0 | 18–23 | 6 | 21.8 | 19–23 | 5 | 22.0 | 19–23 | 5 |
15 | 22.9 | 22–28 | 7 | 21.8 | 20–25 | 6 | 22.0 | 20–25 | 6 |
16 | 22.8 | 19–24 | 6 | 21.8 | 17–23 | 7 | 22.3 | 17–23 | 7 |
17 | 22.8 | 20–23 | 4 | 24.2 | 19–28 | 10 | 23.7 | 19–27 | 10 |
18 | 22.7 | 18–23 | 6 | 20.0 | 17–21 | 5 | 20.1 | 17–22 | 6 |
19 | 22.5 | 20–25 | 6 | 23.2 | 20–29 | 10 | 23.6 | 20–30 | 11 |
20 | 22.4 | 22–30 | 9 | 24.0 | 22–25 | 4 | 23.9 | 22–25 | 4 |
21 | 22.1 | 20–24 | 5 | 21.8 | 19–25 | 7 | 22.0 | 19–25 | 7 |
22 | 22.0 | 18–22 | 5 | 18.7 | 17–19 | 3 | 18.5 | 17–19 | 3 |
23 | 21.9 | 17–22 | 6 | 18.8 | 18–19 | 2 | 19.5 | 18–21 | 4 |
24 | 21.8 | 17–20 | 4 | 20.4 | 18–22 | 5 | 20.3 | 18–22 | 5 |
25 | 21.8 | 18–25 | 8 | 24.2 | 20–29 | 10 | 24.3 | 20–29 | 10 |
26 | 21.5 | 18–20 | 3 | 19.7 | 18–22 | 5 | 19.9 | 18–22 | 5 |
27 | 21.1 | 17–20 | 4 | 19.8 | 17–21 | 5 | 19.8 | 17–21 | 5 |
28 | 21.0 | 18–22 | 5 | 20.0 | 18–21 | 4 | 20.1 | 19–21 | 3 |
29 | 20.8 | 18–22 | 5 | 20.7 | 18–22 | 5 | 20.7 | 18–22 | 5 |
30 | 20.8 | 18–22 | 5 | 20.6 | 19–22 | 4 | 20.3 | 19–21 | 3 |
31 | 20.3 | 18–21 | 4 | 20.2 | 18–22 | 5 | 20.1 | 18–22 | 5 |
32 | 20.0 | 20–24 | 5 | 24.3 | 22–26 | 5 | 24.2 | 22–26 | 5 |
33 | 20.0 | 18–23 | 6 | 20.5 | 18–22 | 5 | 20.6 | 18–22 | 5 |
34 | 20.0 | 16–20 | 5 | 19.5 | 17–20 | 4 | 19.5 | 17–21 | 5 |
35 | 20.0 | 18–24 | 7 | 20.4 | 18–23 | 6 | 20.3 | 18–23 | 6 |
36 | 19.9 | 18–24 | 7 | 22.5 | 17–25 | 9 | 22.5 | 17–25 | 9 |
37 | 19.7 | 17–22 | 6 | 19.9 | 17–22 | 6 | 19.9 | 17–22 | 6 |
38 | 19.6 | 17–20 | 4 | 19.9 | 18–21 | 4 | 20.0 | 17–21 | 5 |
39 | 19.5 | 18–22 | 5 | 19.9 | 17–21 | 5 | 20.0 | 17–21 | 5 |
40 | 19.3 | 18–21 | 4 | 20.2 | 18–22 | 5 | 20.3 | 18–22 | 5 |
41 | 18.6 | 18–22 | 5 | 20.6 | 19–21 | 3 | 20.6 | 19–21 | 3 |
FAR Analysis | McKern and Stewart | Total Sample | |
---|---|---|---|
Age Interval | 8.1 | 6.3 | 6.4 |
Correct | 38/41 | 31/41 | 33/41 |
% Correct | 92.7% | 75.6% | 80.5% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
McCormick, K.A. Comparing Traditional Age Estimation at the Defense POW/MIA Accounting Agency to Age Estimation Using Random Forest Regression. Forensic Sci. 2023, 3, 273-283. https://doi.org/10.3390/forensicsci3020020
McCormick KA. Comparing Traditional Age Estimation at the Defense POW/MIA Accounting Agency to Age Estimation Using Random Forest Regression. Forensic Sciences. 2023; 3(2):273-283. https://doi.org/10.3390/forensicsci3020020
Chicago/Turabian StyleMcCormick, Kyle A. 2023. "Comparing Traditional Age Estimation at the Defense POW/MIA Accounting Agency to Age Estimation Using Random Forest Regression" Forensic Sciences 3, no. 2: 273-283. https://doi.org/10.3390/forensicsci3020020