Etiqueta: biomechanics

Using Machine-Learning tools to infer fossil locomotion from FEA and GMM data

The talus is the most commonly preserved post-cranial element in the platyrrhine fossil record and, consequently, can provide information about postural adaptations because it is the anatomical structure responsible for transmitting body mass forces from the leg to the foot. We published recently a new paper entitled “Inferring locomotor behaviours in Miocene New World monkeys using finite-element analysis, geometric morphometrics and machine-learning classification techniques applied to talar morphology” in the Journal of the Royal Society Interface to test whether the locomotor behaviour of fossil Miocene platyrhines could be inferred from their talus morphology.

One of the important novelties of this paper is that, as I wrote before, when you develop new methods to generate and quantify FEA data, you can use it in new ideas. In this case we are using the brand new methodology of the Intervals’ Methods to infer fossil locomotion combining Finite Element Analysis (FEA) and Geometric Morphomoetrics (GMM) data in Machine-Learning Algorithms.

The extant sample of New-world Monkeys was classified into three different locomotor categories (namely clamber, quadrupedal and leaper) and then talar strength was compared using FEA. GMM were used to quantify talar shape and to assess its association with biomechanical strength. Finally, several machine-learning algorithms were trained using both the biomechanical and morphometric data from the extant taxa to infer the possible locomotor behaviour of the Miocene fossil sample.

The obtained results show that the different locomotor categories are distinguishable using either biomechanical or morphometric data and the machine-learning algorithms categorised most of the fossil sample as arboreal quadrupeds. One of the most important points of this this study is that it has shown that a combined approach can contribute to the understanding of platyrrhine talar morphology and its relationship with locomotion and that machine-Learning tools can also be very useful in fossil taxa.


More info: Thomas A. Püschel; Jordi Marcé-Nogué; Justin T. Gladman; René Bobe; William I. Sellers (2018). “Inferring locomotor behaviours in Miocene New World monkeys using finite element analysis, geometric morphometrics and machine-learning classification techniques applied to talar morphology”. Journal of the Royal Society Interface. 15 (146): 20180520. doi:10.1098/rsif.2018.0520




Parametric analysis in a FEA model to understanding basal urodeles

In vertebrate palaeontology, some previous works joined a Parametric Analysis and FEA to test the behaviour and sensitivity of different parameters such as the material properties of the biological tissue, the homogeneity or heterogeneity of the bone, the sutures, or the influence of the loads applied. Previously we used thiks type of analysis to test how the variation of the original geometry affects the biomechanical performance in the eyes of an Edingerella madagascariensis to study the implications of orbit Position and size diversity of early amphibians:

Now, Our new joined paper “Cranial biomechanics in basal urodeles: the Siberian salamander (Salamandrella keyserlingii) and its evolutionary and developmental implications” with the researchers Zupeng Zhou, Josep Fortuny & Pavel P. Skutschas has been published in Scientific Reports. The work is analysing a 3D cranial biomechanics of the adult Salamandrella keyserlingii  under different tissue properties and ossification sequences of the cranial skeleton to unravel that:

  1. Mechanical properties of tissues (as bone, cartilage or connective tissue) imply a consensus between the stiffness required to perform a function versus the fixation (and displacement) required with the surrounding skeletal elements.
  2. Changes on the ossification pattern, producing fontanelles as a result of bone loss or failure to ossify, represent a trend toward simplification potentially helping to distribute stress through the skull, but may also imply a major destabilization of the skull.
  3. Bone loss may be originated due to biomechanical optimization and potential reduction of developmental costs.
  4. Hynobiids are excellent models for biomechanical reconstruction of extinct early urodeles.


Zhou, Z. et al. “Cranial biomechanics in basal urodeles: the Siberian salamander (Salamandrella keyserlingii) and its evolutionary and developmental implications”. Scientific Reports. 2017. More info at:



Going to the EAVP not only as an attendant

The first week of August I attended the 15th Annual Meeting of the Association of European Vertebrate Palaeontologists (EAVP) in Munich (Germany). I presented my recent join work about biomechanics of the platyrrhine talus with my colleague Thomas Püschel (University of Manchester) where we were studied the biomechanical behaviour of the talus in 40 extant platyrrhine species to test if it was possible to distinguish locomotor behaviours (i.e. suspension, quadrupedalism and leaping). Then, we analysed ten different Miocene fossil samples to compare their stress results with the extant data to infer their possible locomotor behaviours.


But, this was not all my contribution in the conference. Moreover, I was one of the co-organizers of the symposyum “Ecomorphology and functional anatomy in vertebrate palaeontology” together with my colleagues of the Institut Català de Paleontologia (ICP): Josep Fortuny and Soledad De Esteban-Trivigno. This symposium intended to be the meeting point for all functional anatomists and evolutionary ecologists interested in vertebrate palaeontology with more than 20 talks. My talk was included in this symposyum as well the talk of my colleague Josep Fortuny about our join work using Finite Element Analysis (FEA) in current and extanct amphibians and reptiles.

The first day of the conference I was also teaching a four-hours workshop organized by Transmitting Science: “Introduction to Finite Element Analysis”. The workshop introduced vertebrate paleontologists to FEA, which is a great tool to approach problems in biomechanics of living and extinct organisms using digital models.

My works there:

  • Marcé-Nogué, J. [et al.] “Inferring locomotor behaviours in Miocene New World Monkeys: A comparative Finite Element Analysis of the platyrrhine talus” 15th EAVP. Munich (Germany). 2017
  • Fortuny, J. [et al.] “3D Computational Biomechanics Meets Amphibians: Ecomorphology And Evolutionary Implications”. 15th EAVP. Munich (Germany). 2017

And, of course, I took some days off to visit Munich and the most known place in Bayern: The castle of Neuschwanstein!


Feeding biomechanics of Late Triassic Metoposaurids

A new paper has been published in Journal of Anatomy studying the paleoecology of extinct amphibian Temnospondyli. The study aims to expand upon the paleoecological interpretations of these animals using 3D Finite Element Analyses (FEA) because the paleoecology of metoposaurids is controversial; they have been historically considered passive, bottom-dwelling animals, waiting for prey on the bottom of rivers and lakes or they have been suggested to be active mid-water feeders.

Skulls from two taxa, Metoposaurus krasiejowensis, a gigantic taxon from Europe, and Apachesaurus gregorii, a non-gigantic taxon from North America, were analyzed under different biomechanical scenarios. Both 3D models of the skulls were scaled to allow comparisons between them and reveal that the general stress distribution pattern found in both taxa is clearly similar in all scenarios. In light of our results, both previous hypotheses about the paleoecology of these animals can be partly merged: metoposaurids probably were ambush and active predators, but not the top predators of these aquatic environments. To demonstrate that the stress distribution is similar in both scenarios we used the new methodology proposed by Marcé-Nogué et al. 2016 in “Accounting for differences in element size and homogeneity when comparing Finite Element models: Armadillos as a case study” which facilitates the comprehension.


The FEA results demonstrate that they were particularly efficient at bilateral biting, and together with their characteristically anteropositioned orbits, optimal for an ambush strategy. Nonetheless, the results also show that these animals were capable of lateral strikes of the head, suggesting active hunting of prey.

Regarding the important skull size differences between the taxa analyzed, our results suggest that the size reduction in the North American taxon could be related to drastic environmental changes or the increase of competitors. The size reduction might have helped them expand into new ecological niches, but they likely remained fully aquatic, as are all other metoposaurids.

+ more info:

Do you want to learn Finite Element Analysis in the context of biomechanics in life science?


If you want to learn Finite Element Analysis in the context of biomechanics in life science, a new Transmitting Science course is launched for the next July in Catalonia: FINITE ELEMENT ANALYSIS APPLIED TO LIFE SCIENCES

In the course, there is an introduction to the Finite Element in order to model biological structures and understand how they worked. It will cover all the steps involved in FEA except the creation or reconstruction of the model, which it is covered in the previous course Introduction to 3D Imaging Technologies: Photogrammetry, Laser, CT-scan and (µ)CT-scan for Life Sciences by my colleague and teacher of the FEA course, Josep Fortuny. And, I know, sometimes is where we are spending most of our time and efforts!

In the course you will learn how to define the material properties of biological structures, the use of a consistent Mesh Generation Methods, and the proper definition of biomechanical boundary conditions and finally, how understand and analyse the results obtained in plane models (the wrong-called 2D) and in the fancy 3D models created from CT-data. But, the limitation of time is always a problem and, in the course, we are just covering static analysis and linear materials. Which it is enough for an starting course because most of the works published in life sciences are covering this part and creates and open window for everybody in the course to learn more in the future.

You can enroll in the website and ask me questions if you have doubts!

Using statistics in the post-process of FEA results

When we are solving a model using Finite Element Analysis, the results are obtained via a distribution map. These internal distributions of the forces  ̶ called stress ̶  appear in the inner regions of the models due to the action of external forces. To model how different forces act on a biological structure, as a bone for example, computational models are created. These models are subdivided in small pieces called “elements” using a mesh. Then, theoretical forces are applied to the model and the stress values of each element are recorded and mapped in a coloured map called stress distribution which enables a qualitative comparison between different models. Of course, this “coloured map” is related with values. Specifically,. with the values obtained from the solving in each element of the mesh.

Screenshot 2016-09-05 11.32.23.png

To analyse these stress values in a quantitative framework could be complicated, as these elements have different size in the same mesh. for this reason, we published recently the work: ” Accounting for differences in element size and homogeneity when comparing Finite Element models: Armadillos as a case study” in Palaeontologia Electronica. In this work we propose a method to obtain the average mean and median of the distribution of these stresses in a Finite Element model weighting for the differences in elements size. On the other hand we propose a procedure to check whether the meshes used to generate the elements provide accurate results to be used later in statistical analysis. Therefore the stress values can be used as a proxy of the relative strength of vertebrate structures in a comparative framework and allow comparing the obtained mechanical results of different models.

This figure is an example: Box-plots of Von Mises stress distributions when Quasi-Ideal Meshes (QUIM) are assumed for the 20 Cingulata mandibles analysed in the work enabling a qualitative comparison between species and diets.


Citation: Marcé-Nogué, J., Esteban-Trivigno, S. de, Escrig, C., & Gil, L. (2016). Accounting for differences in element size and homogeneity when comparing Finite Element models: Armadillos as a case study. Palaeontologia Electronica, 19(2), 1–22

Read more:


Virtual Paleontology meets China

L’investigador de l’Institut Català de Paleontologia Miquel Crusafont, Josep Fortuny, juntament amb Jordi Marcé-Nogué, investigador de la Universitat d’Hamburg, han estat convidats durant la primera setmana d’abril a la Guilin University of Electronic Technology (ciutat de la Regió Autònoma de Guangxi) a impartir xerrades i treballar conjuntament amb l’equip liderat pel Dr. Zupeng Zhou en l’àmbit de la biomecànica computacional.

Durant aquesta estada, els investigadors catalans i xinesos compartiran coneixements i experiències en una disciplina, la biomecànica computacional, que té el seu origen en l’enginyeria, però que traslladada al camp de la biologia o la paleontologia permet realitzar simulacions biomecàniques tridimensionals en estructures biològiques complexes, tals com cranis o dents, que són difícils o impossibles de dur a terme amb organismes vius.

Aquesta nova col·laboració persegueix obrir aquest camp al continent asiàtic, on fins al moment no hi ha cap equip que uneixi la vessant paleobiològica i d’enginyeria. Gràcies al treball conjunt d’aquests equips de recerca es podran obtenir noves dades molt rellevants, tant sobre els diversos grups d’animals tetràpodes amb els que així com pels nous mitjans tècnics que es desenvoluparan arran d’aquesta nova col·laboració.

L’objectiu de la biomecànica computacional és reproduir el moviment i les tensions que es produeixen en una determinada estructura d’una forma no invasiva. Aquestes anàlisi són d’especial utilitat en especies extingides on no es poden observar aquests comportaments en viu, així com també per conèixer la biologia i ecologia d’espècies poc conegudes o bé en perill d’extinció. En aquests casos, conèixer la seva biologia pot ser de gran utilitat per establir mesures correctores.

El Departament de Paleontologia Virtual de l’ICP fa anys que treballa en aquest camp, unint paleontologia, biologia i enginyeria i col·laborant amb altres institucions com la universitat Politècnica de Catalunya o la Universitat d’Hamburg. Es tracta d’una unitat transversal als altres grups de recerca del centre que utilitza tecnologies no invasives en l’estudi dels fòssils per tal d’explorar i quantificar estructures habitualment no visibles. Les tècniques inclouen modelització 3D, tomografia computada industrial i mèdica, escaneig làser i fotogrametria així com tècniques d’enginyeria com l’Anàlisi d’Elements Finits.


(text: Pere Figuerola, ICP)