Contents

MSc | PhD | PostDoc Positions

At the NeuroPoly lab at Polytechnique & Université de Montréal (www.neuro.polymtl.ca), best student-city in the world! (https://www.topuniversities.com/city-rankings/2017), we develop advanced MRI image analysis techniques using deep learning and distribute them as open-source software. In collaboration with neuroradiologists and world experts in deep learning (Mila), we apply these tools in patients with traumatic injury and neurodegenative diseases (multiple sclerosis, ALS, etc.).

We are recruiting Master/PhD students and Postdoc fellows to work on various projects:Edit

Deep Learning for Medical Applications (in partnership with Mila) ⚙️⚙️

  • Multiple sclerosis (MS) lesion segmentation on MRI

    • Description: Segment MS lesions in order to help classify these patients (phenotype, personalized therapy). Dataset: MRIs of ~650 patients from 12 international clinical centers, MS lesions labeled by neuroradiologists. Methods: Develop/apply advanced deep learning architectures (FiLM, HEMIS, etc.).

    • Skills: Deep learning | Image analysis | Python | git/GitHub

    • Related to: ivadomed | Spinal Cord Toolbox

  • Segmentation of axons and myelin from histology

    • Description: Segment axons and myelin sheath (two different labels) from large-scale histology data, in order to create microstructure atlas of the human central nervous system (example: see BigBrain project: https://bigbrain.loris.ca). Dataset: Electron microscopy and optical imaging, with about 15,000 axons and myelin sheath manually segmented.

    • Skills: Deep learning | Image analysis | Python | git/GitHub

    • Related to: AxonDeepSeg | ivadomed

  • Generalization of deep learning models

    • Incorporation of prior constraints from MRI physics during training

    • Synthesizing images during data augmentation

  • Continual learning

    • Continual learning strategies when data sharing is restricted;

    • Implementing continual learning strategies in neuroimaging and clinical centres

  • Uncertainty estimation and applications

    • Incorporating uncertainty measures to deal with inter-rater variability

    • Using uncertainty in active learning framework for histology and medical data segmentation

    • Use DeepSeg method to encode uncertainty.

  • Implementing open-source AI solutions

    • Bringing AI methods into clinical radiology routine via user-friendly software solutions;

    • Contribution to medical AI framework ivadomed

Edit

Neuroimaging Analysis 🧠

  • Pipelines for processing large neuroimaging datasets

    • Description: Set up processing pipelines for analysis large databases of patients. Datasets: UK Biobank, private databases from collaborating hospitals.

    • Skills: Image analysis | Neuroimaging tools | Programming | git/GitHub

    • Related to: Spinal Cord Toolbox

MRI Physics 🧲

  • Realtime shimming with MRI

    • Description: Building on our recent progress on integrated shim coils and real time shimming technology, we are recruiting Master/PhD/Postdoc fellows to work on real time shimming projects applied to the spinal cord at 7T. Research will be conducted at the NeuroPoly lab (Polytechnique, University of Montreal, www.neuro.polymtl.ca), and at the Montreal Neurological Institute (MNI, McGill University).

    • Skills: Ultra-high field MRI | MRI acquisition | Image analysis |

    • Related to: Shimming Toolbox

    • More details here

What profile are we looking for?

  • For projects on image analysis | deep learning:

    • Strong coding skills in Python, use of git/GitHub

    • Passion for open-source software, data science and knowledge sharing ❤️

    • Experience in data science, computer vision and medical imaging is an asset

  • For projects on MRI physics:

    • Strong theoretical background in physics

    • Experience in MRI

    • Coding skills in Python/Matlab, use of git/GitHub

Why you should join us?

  • Join an environment that fosters autonomy, passion and creativity;

  • Get the opportunity to take leadership in open-source projects with strong impact in the medical field;

  • You will gain highly relevant expertise on bleeding edge technologies (MRI physics, computer vision, A.I., deep learning, etc.)

  • You will be using a state-of-the-art 7T Siemens Terra system

  • You will interact with radiologists and neurosurgeons who will apply these techniques

  • You will collaborate with top institutions (Univ. Montreal, McGill, MGH/Harvard, UCL, Oxford, etc.)

How to apply?

  • Send requests to Julien Cohen-Adad (include CV, GitHub link, blogs, grades, references)