Section: Partnerships and Cooperations

International Initiatives

Inria Associate Teams Not Involved in an Inria International Lab

Snowball

• Title: Discovering knowledge on drug response variablity by mining electronic health records

• International Partner (Institution - Laboratory - Researcher):

  • Stanford (United States) - Department of Medicine, Stanford Center for Biomedical Informatics Research (BMIR) - Nigam Shah

• Start year: 2017

• See also: http://snowball.loria.fr/

Snowball (2017-2019) is an Inria Associate Team and the continuation of the preceding Associate Team called Snowflake (2014-2016). The objective of Snowball is to study drug response variability through the lens of Electronic Health Records (EHRs) data. This is motivated by the fact that many factors, genetic as well as environmental, imply different responses from people to the same drug. The mining of EHRs can bring substantial elements for understanding and explaining drug response variability.

Accordingly the objectives of Snowball are to identify in EHR repositories groups of patients which are responding differently to similar treatments, and then to characterize these groups and predict patient drug sensitivity. These objectives are complementary to those of the PractiKPharma ANR project. Moreover, it should be noticed that Adrien Coulet has started a one-year sabbatical stay in the lab of Nigam Shah at Stanford University since September 2017.

Participation in International Programs

A stay at NASA Frontier Development Lab

In July 2017, Chedy Raïssi visited NASA Ames and SETI Institute as part of the Frontier Development Lab. He worked on mentoring teams and developing meaningful research opportunities, as well as supporting the work of the planetary defense community and showing the potential of this kind of applied research methodology to deliver breakthrough of significant value.

Delay-Doppler radar imaging is a powerful technique to characterize the trajectories, shapes, and spin states of near-Earth asteroids and has yielded detailed models of dozens of objects. Since the 1990s, delay-Doppler data has been analyzed using the SHAPE software developed originally by Steven J. Ostro. SHAPE performs sequential single-parameter fitting, and requires considerable computation runtime and human intervention. Recently, multiple-parameter fitting algorithms have been shown to more efficiently invert delay-Doppler datasets thus decreasing runtime while improving accuracy. However, reconstructing asteroid shapes and spins from delay-Doppler data is, like many inverse problems, computationally intensive and requires extensive human oversight of the shape modeling process.

Thus we have explored two new techniques to better automate delay-Doppler shape modeling: Bayesian optimization and deep generative models. Firstly we have implemented a Bayesian optimization routine that uses SHAPE to autonomously search the space of spin-state parameters. Bayesian optimization yielded similar spin state constraints with computer runtime reduced by a factor of 3. Secondly, the shape modeling process could be further accelerated using a deep “generative model" to replace or complete iterative fitting. Accordingly, we have implemented and trained a deep generative model based on different architectures of deep convolutional networks. Results are currently under analysis and future publications are in preparation.

LEA STRUCO

Participant : Jean-Sébastien Sereni.

LEA STRUCO is an “Associated International Laboratory” of CNRS between IÚUK, Prague, and LIAFA, Paris. It focuses on high-level study of fundamental combinatorial objects, with a particular emphasis on comprehending and disseminating the state-of-the-art theories and techniques developed. The obtained insights shall be applied to obtain new results on existing problems as well as to identify directions and questions for future work. Jean-Sébastien Sereni is the founder and previous director of LEA STRUCO, which was initiated when Jean-Sébastien was a member of LIAFA, and he is now a member of its scientific committee.

Research Collaboration with HSE Moscow

Participants : Miguel Couceiro, Adrien Coulet, Tatiana Makhalova, Amedeo Napoli, Chedy Raïssi, Justine Reynaud.

An on-going collaboration involves the Orpailleur team and Sergei O. Kuznetsov at Higher School of Economics in Moscow (HSE). Amedeo Napoli visited HSE laboratory several times while Sergei O. Kuznetsov visits Inria Nancy Grand Est every year. The collaboration is materialized by the joint supervision of students (such as the thesis of Aleksey Buzmakov defended in 2015 and the on-going thesis of Tatiana Makhalova), and the the organization of scientific events, as the workshop FCA4AI with five editions between 2012 and 2016 (see http://www.fca4ai.hse.ru).

This year, we participated in the organization of two main events: a special session about Knowledge Discovery and Formal Concept Analysis at the ISMIS Conference in Warsaw (Poland) in June 2017 (http://ismis2017.ii.pw.edu.pl/s_kd_fca.php), and the chairing of the track “General Topics of Data Analysis” at the AIST Conference in Moscow in July 2017 (6th International Conference on Analysis of Images, Social Networks, and Texts http://aistconf.org/). Finally a next edition of the seventh edition of the FCA4AI workshop is planned in July 2018 at the ECAI-IJCAI Conference to be held in Stockholm Sweden.