Navigation – Plan du site

AccueilLes journées de la SAP36(S)Résumés des journées 2024The 'Reverse Engineering' approac...

Résumés des journées 2024

The 'Reverse Engineering' approach to hominin long bone reconstruction: Transforming human evolution studies

La méthode de 'rétro-ingénierie' pour la reconstruction des os longs d’hominines : impact sur les études de l’évolution humaine
Lloyd Austin Courtenay et Julia Aramendi

Texte intégral

1Our grasp of human evolution faces limitations due to the challenges posed by the discovery of isolated and often incomplete hominin remains in a complex evolutionary context. Over time, paleoanthropologists have developed various techniques for reconstructing these fragments, which can sometimes introduce methodological uncertainties and biases. In this context, we introduce a pioneering approach known as the 'Reverse Engineering' method, which focuses on reconstructing long bones of hominins. We illustrate this methodology using a case study involving the reconstruction of fragmented humeri, radii, femora, and tibiae from Homo naledi. This approach integrates 3D geometric morphometrics and advanced computational, mathematical, and artificial intelligence tools, utilizing complete long bones from modern human and primate reference groups. By combining landmarks and semilandmarks, we have created a database describing morphological variations among anatomically modern humans, chimpanzees, gorillas, and orangutans. For the H. naledi specimens, we initially aligned and mirrored them to establish correspondences with complete modern long bones before landmarking. Subsequently, we employed the 'Reverse Engineering' method to estimate missing landmarks, using geometric morphometric information processed through dimensionality reduction techniques. This enabled us to establish a mathematical relationship between preserved bone portions in each fossil fragment and complete modern long bones, ultimately predicting the morphology of the entire bone corresponding to each fossil fragment. Our pursuit of this mathematical relationship included experimenting with multiple different regression algorithms. Neural Networks trained using meta-learning approaches were found to obtain the best accuracy, with reported improvements of 3.8mm compared to other methods. This innovative approach successfully reconstructed all available H. naledi long bone specimens. These finds carry substantial implications for the study of human evolution, particularly for fragmentary hominin fossils.

Haut de page

Pour citer cet article

Référence électronique

Lloyd Austin Courtenay et Julia Aramendi, « The 'Reverse Engineering' approach to hominin long bone reconstruction: Transforming human evolution studies »Bulletins et mémoires de la Société d’Anthropologie de Paris [En ligne], 36(S) | 2024, mis en ligne le 22 décembre 2023, consulté le 21 février 2024. URL : ; DOI :

Haut de page


Lloyd Austin Courtenay

PACEA UMR 5199, CNRS, Université de Bordeaux, Ministère de la Culture, Pessac, France ; ladc1995[at]

Julia Aramendi

McDonald Institute for Archaeological Research, University of Cambridge, Cambridge, United Kingdom

Articles du même auteur

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search