Nina Miolane

I am a postdoctoral researcher and lecturer at Stanford University (USA), in the Statistics Department and in collaboration with the Psychology Department of Stanford University (USA) and the Project-Team Epione of Inria (France). I work with Susan Holmes, Russell Poldrack and Xavier Pennec.

My research aims to create an ontology of human organ shapes, i. e. a systematic and quantitative description of organ shapes and their variations in the population. Eventually, this ontology will evolve towards a description of both shapes and functions of organs, by bringing in new image modalities and heterogeneous biological data. 

Along the way, I am passionate about building the mathematical theory to support the above medical application. I am contributing to the development of "Geometric Statistics" which are statistics for data belonging to non-Euclidean spaces, just like shapes belong to non-Euclidean spaces. I am devoted to providing general algorithms that are applicable not only to organ shapes, but to a broader range of shape data, as well as to data belonging to general non-Euclidean spaces like Riemannian manifolds, Lie groups or quotient spaces. All in all, this research is a very exciting pluridisciplinar enterprise blending Mathematics, Computer Science, Statistics and Medical Imaging for applications in Medicine.

Prior to this, I graduated from my PhD at Inria Sophia Antipolis (France) on computational anatomy. I also worked as a deep learning software engineer for the San Francisco based startup Bay Labs, on automatic acquisition and analysis of echocardiographies.

Contact information

  • Addresses:
Inria, Epione Project-Team, Office F224, Bâtiment Fermat, 2004 route des lucioles, 06902 Sophia-Antipolis, France

Stanford University, Department of Statistics, Sequoia Hall, 390 Serra Mall, CA 94305 Stanford, United States

Bay Labs, 290 King Street, 94107 San Francisco, United States
  • Emails: nmiolane at; nina.miolane at; nina at


Journals with peer-reviews
Conferences with peer-reviews

Ph.D Thesis

  • Python package geomstats: open-source package for computations of Riemannian geometry in Machine Learning. See examples to start using it. Contact us to contribute!


Research interests

  • Neuroscience
  • Computational Medicine
  • Applied Statistics and Machine learning,
  • Riemannian, pseudo-Riemannian and sub-Riemannian geometry,
  • (Organ) shape analysis,
  • (Medical) Image processing,
  • (Medical) Image Analysis.


  • Course Stats 110: Statistical Methods for Engineering and the Physical Sciences. Stanford Summer and Fall 2018.


  • NeuroImage (2014)
  • Journal of Mathematical Imaging and Vision (2015, 2017)
  • NIPS conference (2016, 2018)
  • GSI conference (2017)
  • Journal Biometrika (2018)

Member of scientific committees




  • PhD in "Geometric Statistics for Computational Anatomy". Inria, France, in collaboration with Stanford, USA.
  • MS. in Theoretical Physics "Quantum Fields and Fundamental Forces". Imperial College London, UK.
  • "Diplôme d'Ingénieur" and MS. in Mathematics and Theoretical Physics. Ecole Polytechnique, France.






  • One of the lead organizers of the RedX Lectures, every Tuesday at the Red Victorian, San Francisco (2017-2018).
For general audience
  • Public talk for the “Galactic Tick Day” Event. Symmetry Labs, San Francisco (2018).
  • Public talk. What do General Relativity and Brain Shape Analysis Have in Common? Red Victorian, San Francisco (2018).
  • Pro-speaker at the European Youth Debating Competition. (2017).
  • Public talk. Annual workshop of Art-Cerveau-Pensée association. Médecine numérique et incertitude. (2017).
  • NVIDIA webinar. How AI is transforming Healthcare. (2017).
  • Public talk, Ma thèse en 180 secondes (3 Minute Thesis). "Le diagnostic, demain". Video, in French. (2017).
  • Panelist at Inria@SiliconValley conference. Digital Health. (2016).
For kids

Communication on science and scientists

In highschools:

Invited talks

  • 2018, July 15th - Stockholm, Sweden. International Conference of Machine Learning (ICML), Workshop Geometry in Machine Learning (GiMLi). Geometric Statistics: Learning from Medical Images?
  • 2018, May 19th - Baltimore, USA. Seminar of John Hopkins University, Center of Imaging Science. Geometric Statistics for Image Analysis.
  • 2018, January 28th - Oberwolfach, Germany. Meeting of the Mathematisches Forschungsinstitut Oberwolfach. Geometric Statistics for Computational Anatomy in the Session on Statistics for Data with Geometric Structures.
  • 2017, December 20th - Goettingen, Germany. Goettingen Statistics Seminar. Open questions in Geometric Statistics.
  • 2017, December 16th - London, UK. 10th International Conference on Computational and Methodological Statistics (CMStatistics 2017). Geometric Statistics for Computational Anatomy in the Session on Statistics for data with geometric structure.
  • 2017, December 13th - London, UK. BioMedIA Seminar of the Department of Computing, Imperial College London. Geometric Statistics for medical image computing.
  • 2017, November 29th - Kingston, Canada. Seminar at Queen's Hospital. AI for Diagnostic Radiology.
  • 2017, November 16th - Cambridge, UK. Shape analysis and computational anatomy workshop. Template shape estimation: correcting an asymptotic bias.
  • 2017, September 23rd - Orlando, USA. American Mathematical Society (AMS) Sectional Meeting. Estimation on manifolds: synchronization of rotations for cryo-electron microscopy in the Special Session on Mathematics of Biomolecules: Discrete, Algebraic, and Topological.
  • 2016, June 7th - Stanford, USA. Stanford Statistics Seminar. Template shape estimation in Computational Anatomy.
  • 2015, November 2nd - Montpellier, France. Statistics Seminar of Université de Montpellier. Statistical properties of the Fréchet mean in quotient spaces. Applications to Computational Anatomy.
  • 2015, February 19th - Vienna, Austria. International Workshop on Infinite-Dimensional Riemannian Geometry with Applications to Image Matching and Shape Analysis. Erwin Schroedinger International Institute for Mathematics and Physics. Noise Effects on Quotient Spaces M/G.

Useful links