[robotics-worldwide] [jobs] PhD Position in Computer Vision and Medical Imaging

TAMADAZTE brahim.tamadazte at femto-st.fr
Mon Jul 15 01:20:11 PDT 2019

PropositionPhD position Title Registration of 3D optical biopsies:
application to medical imaging of high-resolution and wide
field-of-viewLocation •	FEMTO-ST Institute (UMR 6174), AS2M Department•
25000, Besançon, FranceDuration•	36 months Scientific context FEMTO-ST
Institute (Besançon, France), develop several research activities related to
surgical robotics, medical imaging, and image-guided intracorporeal
interventions. In particular, we are addressing the clinical problem of
performing of in-situ 3D optical biopsies, i.e. the optical analysis at the
micrometer resolution of biological tissues thanks to highly efficient
medical imaging without the traditional physical sampling of suspected
tissues (physical biopsy). However, tissue analysis at micrometric
resolution, using for example optical coherence tomography (OCT) or confocal
microscopy, raises the problem of the limited field of view. Actually, this
is the recurring dilemma of a wide field of view and micrometric resolution.
To remedy this, it is necessary to perform several high-resolution 3D
optical biopsies based on the micrometric motion of the imaging probe and to
assemble them to reconstruct a single high-resolution biopsy over a large
area. Clinical contextThe clinical objectives of this investigation are
numerous, among them, achieving of 3D optical biopsies with micrometric
resolution and wide field of view as part of the in-situ characterization of
the olfactory mucosa in human subjects. The idea to find a precursor sign
(earlier diagnosis) of a future neurodegenerative disease (Alzheimer’s and
Parkinson’s disease) by studying the odor operating in its healthy and
pathological states.  The diagnosis is intended to be as early as possible
(5 to 10 years) before the first signs of motor impairment or memory loss.
In fact, it has been statistically demonstrated (post-mortem studies) that
loss of smell with no apparent causes (trauma, cancer, etc.) is generally
correlated with dementia (neurodegenerative diseases, such as Alzheimer).
Consequently, the aim is to understand what is happening in vivo in the
olfactory mucosa without taking a sampling process of tissues (physical
biopsy), which is clinically difficult to achieve. However, the in-situ
characterization of the olfactory mucosa can be carried out thanks to 3D
optical biopsies approaches. Scientific content The scientific work concerns
the development of accurate and robust registration methods between 3D
visual information (optical biopsies). The PhD student will first focus on
the development of calibration methods based on local visual features (e.g.,
geometric type information), matching methods, computing rigid
transformations as well as reconstruction and mapping all the acquired data
on the same final volume. This first work will provide a basis for extending
these methods in aim to use global information such as pixel/voxel
intensities, wavelet coefficients or any other global image information. In
addition, several parameters and external disturbances will be considered
within this work such as physiological motion, … in order to limit potential
artifacts. This work will then be extended to non-rigid registration in
which the observed tissue is subjected to a global (or a local minimum)
deformation resulting, for instance, from the contact between the tissue and
the imaging probe. Required knowledges •	(medical) imaging, computer vision,
robotics and applied mathematics •	Advanced programing skills (C++, Matlab,
Python)•	Knowledges of computer vision libraries (OpenCV, PCL) will be
appreciatedSupervision team •	Guillaume Laurent (Associate-Prof, ENSMM)    -
https://urldefense.proofpoint.com/v2/url?u=https-3A__gjlaurent.github.io_&d=DwIFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rSs8MCeFamFEPafDTOhgTfYI&m=v0WNmecimhOHobNYGjRVjXC_G8dpl6JcUaFPm63pQww&s=AIbAhJh786XCTWuzNZJD7DG6inr4SAOcIxkkhyjqwTE&e= •	Brahim Tamadazte (Senior Scientist, CNRS)    -
https://urldefense.proofpoint.com/v2/url?u=https-3A__sites.google.com_site_tamadazte_home&d=DwIFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rSs8MCeFamFEPafDTOhgTfYI&m=v0WNmecimhOHobNYGjRVjXC_G8dpl6JcUaFPm63pQww&s=wg-u7483-WFchfhsZUUb_sIYso1Oltdh97-G8vd09TM&e=  Application procedureInterested
candidates should send a cover letter, a CV and transcripts (M1 and M2 or
equivalent) as a single PDF file to the supervisors before 1 September
2019.References [1] Rosa, B., Dahroug, B., Tamadazte, B., Rabenorosoa, K.,
Rougeot, P., Andreff, N., & Renaud, P. (2018). Online robust endomicroscopy
video mosaicking using robot prior. IEEE Robotics and Automation Letters,
3(4), 4163-4170.[2] Ourak, M., Tamadazte, B., & Andreff, N. Partitioned
camera-OCT based 6 DOF visual servoing for automatic repetitive optical
biopsies. In 2016 IEEE/RSJ International Conference on Intelligent Robots
and Systems (IROS) (pp. 2337-2342). 2016.[3] Ourak, Mouloud, Tamadazte,
Brahim, Laurent, Guillaume J., et al. Geometric Calibration of an OCT
Imaging System. In: 2018 IEEE International Conference on Robotics and
Automation (ICRA), 2018. p. 3993-3999.

Sent from: https://urldefense.proofpoint.com/v2/url?u=http-3A__robotics-2Dworldwide.1046236.n5.nabble.com_&d=DwIFaQ&c=clK7kQUTWtAVEOVIgvi0NU5BOUHhpN0H8p7CSfnc_gI&r=0w3solp5fswiyWF2RL6rSs8MCeFamFEPafDTOhgTfYI&m=v0WNmecimhOHobNYGjRVjXC_G8dpl6JcUaFPm63pQww&s=fWd3SJW4RNB9kCt2nJzSrZpqnf8EAm_6QCuf_8RQX1U&e= 

More information about the robotics-worldwide mailing list