Infer Fossil Behaviour from Biomechanical and Morphological Data Using Machine Learning

Morphological and biomechanical methods have been broadly used in the study of species evolution in the last twenty years. Nowadays, computational methods from engineering such as geometric morphometrics, computational fluid dynamics, kinematical models and finite element analysis are common in the tool-kit of evolutionary studies. Specifically, computational mechanics by means of finite element analysis (FEA) has been conducted in a wide spectrum of vertebrate bony structures providing new insights into the functional constraints and the adaptive value of different morphologies. Recently, some powerful and innovative approaches were proposed to post-process FEA outputs in a quantitative manner that were specifically designed to compare several different models using multivariate statistics and machine-learning algorithms. The application of this methods successfully showed that it is possible to effectively discern dietary preferences and locomotor behaviours when analysing biomechanical data. Moreover, the use of machine-learning algorithms allowed to advance dietary and locomotor classifications for fossil species where this information remains unknown. These new methods from engineering are opening a new window to introduce biomechanical models into an evolutionary framework providing a new set of tools and data to the paleobiology and evolutionary scientific communities.


març 31 2023


12:00 - 13:00


Sala de Graus, ETSEQ


  • Jordi Marcé Nogué
    Jordi Marcé Nogué
    Mechanical Engineering Department, URV

    Jordi Marcé-Nogué is an engineer who focused his research in computational mechanics and who specialized in complex and non-linear biomechanical simulations using finite element analysis (FEA) with a clear multidisciplinary vocation. He directs his research to questions important to the field of paleobiology and anthropology and is an internationally recognized by using computational biomechanics in evolution combined with machine learning methods. He has worked in the Universitat Politècnica de Catalunya, in the Center of Natural History of the University of Hamburg and as a researcher scientist in the Department of Pathology and Anatomical Sciences at the Jacobs School of Medicine and Biomedical Sciences of the University at Buffalo, State University of New York. Nowadays he is an Associate Professor in the Departament d’Enginyeria Mecànica of the Universitat Rovira i Virgili where he develops his academic career and his research teaching in mechanical engineering courses.