French
English
Contacts
- Thomas Grenier, Creatis – team Image & Model : thomas[dot]grenier[at]creatis[dot]insa-lyon[dot]fr
- Michael Sdïka, Creatis – team Image & Model : michael[dot]sdika[at]creatis[dot]insa-lyon[dot]fr
- François Cotton, Creatis – team NMR & Optics : francois[dot]cotton[at]chu-lyon[dot]fr
- Application deadline : January 2019
- Beginning of internship : February to end of March 2019
- Internship duration : 4 to 6 month
- Financial support : 540€ / month
- Internship location : CREATIS, Campus de la Doua, Lyon, France
Keywords:
Deep learning detection and segmentation, MRI, medical image processing, 3D analysis.
Project : Develop and test a 3D robust lesion detection approach of active lesions
The internship focuses on the design of an approach able to efficiently detect/segment multiple
sclerosis active lesions from MR images acquired before gadolinium injection. Actual clinical MRI
protocols studying multiple sclerosis follow up are based on a controversial usage of Gadolinium (Gd).
Such usage allows distinguishing precisely the active lesions from the others which then permits
pharmaceutical treatment modifications. The first step is to detect and segment all MS lesions and
then identify active ones. The final objective is to perform active lesions detection but without using
MR images acquired after injection of Gd. To do so, we can exploit deeply the available MRI modalities
acquired before the injection. The ones acquired post injection can be used to create the ground truth
used for training. Thus one main challenge is to propose a novel DNN architecture that exploits
optimally the different MR modalities and the 223 available patient’s MRIs.
Context within CREATIS laboratory
CREATIS is a biomedical imaging research laboratory, with about 200 persons, whose main areas of
excellence and international influence are linked to the identification of i) major health issues that can
be addressed by imaging and ii) of theoretical barriers in biomedical imaging related to signal and
image processing, modelling and numerical simulation.
CREATIS meets these challenges through a multidisciplinary approach, based on a matrix organisation
which stimulates interaction between six research teams working in information and communication
science and technology, engineering sciences and life sciences.
The CREATIS team « Image and Models » explores new medical images analysis approaches in order to
improve knowledge and understanding of diseases or computer aided diagnosis tools. This team
performs upstream research, yielding the design of advanced image processing and modelling
methods such as Deep Learning approaches.
The “NMR and Optics” team goal is to develop new ways of measuring MR-based indirect parameters
and in the process, to seek for new biomarkers. This team has experience and knowledge in many
applied magnetic resonance aspects going from theory of MR physics to medical validation and
applications. Bolstered by its experience and taking into account that Magnetic Resonance is a
modality that often plays a central role for many biomedical imaging investigations, the team
members are involved in most of transversal, inter-team projects.
The proposed project involves researchers of “image & model” and “NMR & Optics” teams and is also
a first milestone of a national project submitted to ANR which aim at studying the usefulness and MR
alternatives of Gd injection in multiple sclerosis follow up. If accepted, this national project will involve
two others French laboratories and will propose PhD opportunities.
Initial skills
- Programming: C++/python/bash, – library : ITK, TensorFlow, PyTorch.
- Image processing: filtering and noise removal, segmentation, machine learning
- Medical imaging : MRI
Expected applicant profile
University Master or Engineering School student (last year of study) with computer science, image analysis and/or applied mathematics profile
Interest, curiosity, learning capability and creativity are appreciate qualities,
Positive spirit, communication skills and ability to work in a team is necessary,
Autonomy, dynamism and motivation to advance his/her own part of the project,
Excellent methodological and hands-on computer programming skills
Facility of understanding and manipulating mathematical models.
Student considering to continue as PhD student (we are searching funds for a PhD on this project)