Postdoc M/F – Reconstruction of Dental CT Scans Using Machine Learning

Organisation/Company CNRS Department Centre de Recherche En Acquisition et Traitement de l'Image pour la Santé Research Field Engineering Computer science Mathematics Researcher Profile Recognised Researcher (R2) Application Deadline 16 Nov 2026 - 23:59 (UTC) Country France Type of Contract Temporary Job Status Full-time Hours Per Week 35 Offer Starting Date 1 Dec 2026 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

This position is part of the collaborative "Cone-Beam AI" project between Acteon, a company specializing in the design of X-ray-based medical devices for the dental sector, the research center for image acquisition and processing in healthcare (CREATIS), and the Institute of Mathematics of Bordeaux (IMB). The main objective of this research project is to improve the quality of cone-beam computed tomography (CBCT) scans used in dental imaging for diagnostic purposes, while reducing the radiation dose received by patients. To achieve this, the postdoctoral researcher will be tasked with developing new, higher-performance tomographic reconstruction algorithms, in particular using machine learning methods suited to solving inverse problems.

In collaboration with the other project members, the postdoctoral researcher's activities will focus on the following topics:

• CBCT reconstruction from truncated data: the goal is to achieve quantitative reconstruction within the scanner's field of view, and the best possible reconstruction outside of it.
• Motion correction: involuntary movements of the patient or the scanner during acquisition introduce artifacts that can be corrected using motion estimation. The objective here will be to estimate this motion without relying on a prior reconstruction, using mathematical equations known as data consistency conditions.
• Scatter radiation: CBCT reconstruction assumes that scattered radiation is not measured, which is not the case in practice and significantly degrades CBCT images. The objective will be to correct X-ray projections by estimating this scattered radiation through a deep learning-based approach.
• Metal artifacts: metals within the field of view cause significant image quality degradation that can compromise diagnosis; these could also be corrected in the projections using deep learning.

Supervisor: Simon Rit
Location: primarily at the Léon Bérard Cancer Centre, Lyon, France.
Duration: 14 months, starting as soon as possible.

CREATIS is a multidisciplinary laboratory with a broad range of expertise in medical imaging, playing a major role in the field of health technologies. The successful candidate will join the tomoradio team, in particular the group based at the Léon Bérard Cancer Centre.

Academic background: PhD in applied mathematics, computer science, or medical physics,
Scientific interests: applied mathematics, computer science (medical image processing), medical physics,
Programming skills: Python, C++ (ITK, RTK),
Languages: English required, French optional.

#J-18808-Ljbffr