Internship - Image Reconstruction for Multi-Source Spectral Medical Scanners - F/H

Let's build a future of trust together

Thales is a global leader in high technologies specializing in three sectors: Defense & Security, Aeronautics & Space, and Cyber & Digital. It develops products and solutions that contribute to a safer, more environmentally friendly, and more inclusive world. The Group invests nearly 4 billion euros per year in Research & Development, particularly in key innovation areas such as AI, cybersecurity, quantum, cloud technologies, and 6G. Thales has nearly 81,000 employees in 68 countries.

Our commitments, your benefits

• Our technological expertise
• Our attention to employee work-life balance
• An inclusive and benevolent environment
• Recognized societal and environmental commitment (Thales Solidarity, CAC 40 ESG index…)

Your daily life

Come discover the site, a human-sized site with a very beautiful environment. Coming to work here means: a friendly and warm atmosphere, interesting engineering projects, mountains nearby.

CONTEXT:

You will carry out this internship within the Research and Development department, specialized in radiological image processing and airport security. You will join the Innovation team, responsible for preliminary projects for Radiology. In recent years, Thales has developed multi-source scanner technology offering several advantages, notably increased compactness and lightness. The internship subject is part of a research project aiming to explore opportunities related to new spectral scanners based on this multi-source architecture.

The objective of the internship is to design 3D image reconstruction algorithms adapted to this type of scanner. Quantitative information extracted from these algorithms — for example, the measurement of the concentration of a contrast agent or the improvement of image quality — can contribute to medical diagnosis.

YOUR MISSIONS:

To obtain the output image of an X-ray scanner, it is necessary to use a reconstruction algorithm to reconstruct a 3D image (e.g., of the patient) from the data acquired by the scanner. The design of efficient reconstruction algorithms must take into account the physics of the detector and, more broadly, the parameters of the acquisition system (system geometry(ies), type of X-ray sources, type of detector, etc.).

The goal of this internship is to develop reconstruction algorithms specifically adapted to multi-source spectral scanners.

To extract quantitative information (e.g., to discriminate soft tissues and a contrast agent), conventional methods use a two-step approach: first, a component decomposition (soft tissues / contrast agent) from spectral information, then a 3D reconstruction. These approaches, although widely used, have limitations, particularly in terms of amplification of acquisition noise, and they are not optimized for multi-source architectures.

In recent years, so-called “one-step” algorithms have shown interest in terms of image quality, as they are often more robust to noise and better suited to the acquisition geometries proposed by multi-source spectral scanners. The objective of the internship is to explore and evaluate these “one-step” approaches for 3D image reconstruction on multi-source scanners.

Your missions:

• Realize a bibliographic review on two-step and one-step methods for spectral reconstruction and their adaptations to multi-source geometries.
• Design, develop, and implement 3D reconstruction algorithms specific to multi-source spectral scanners.
• Compare the performance of the developed algorithms with conventional methods (image quality, noise amplification, robustness, tissue discrimination capability).
• Evaluate the feasibility, advantages, and limitations of the studied approaches in a medical application context.
• Document the work (reports, code) and present the results during team meetings.

You will benefit from the technical support of the Innovation team for Radiology and will have realistic data to work in conditions close to the field.

YOUR PROFILE:

Do you have the ambition to develop new skills within a large international group?

Do you want to discover a rich and varied profession?

Are analysis, autonomy, proactivity, and teamwork recognized strengths of yours?

Are you preparing for a BAC+5 in Applied Mathematics / Computer Science, Image Processing and looking for an end-of-studies internship?

You have knowledge in:

• Applied Mathematics: convex optimization, probabilities, statistics, inverse problems.
• Image Processing: image reconstruction, denoising, etc.
• Programming: Python, C++.
• English: sufficient level to read and understand scientific articles.

Then this position is for you!

All our internships are formalized by an agreement and subject to remuneration, the amount of which is determined according to your level of study.

Thales, a Handi-Engaged company, recognizes all talents. Diversity is our best asset. Apply and join us!