Phd - Computer-assisted guidance of radio-embolization procedures in the context of interventional

Radioembolization (also known as selective internal radiation therapy – SIRT) is a minimally invasive procedure that combines embolization and radiation therapy, where radioactive particles are delivered to a tumor through the bloodstream. This treatment permits a high dose of radiation to be focused directly inside tumors; generally, nearby healthy tissue receives minimal radiation. It is generally an option for some patients with tumors that cannot be removed by surgery, especially for liver cancer, and for cancers whose spread is limited to the liver. It is also often used in coordination with more well-established cancer treatments, such as surgery and chemotherapy. In terms of cost, it is now accepted that this treatment strategy is better for both the patient and the society.

Problematic:

Even if globally more efficient than other therapies, this methodology can still be optimized in terms of time and cost, as well as personalization. Without additional computer-assisted planning and guidance tools the technique remains very dependent of the level of expertise of the operator. For instance, the ideal injection point is chosen by the radiologist based on a detailed portrait of the arterial tree captured by medical imaging. However, without any computer-assisted help this choice is unreproducible and subject to human error. A computer-assisted planning tool based on precise multi-modality segmentation and registration, as well as intra-operative fusion capabilities between the different source of imaging would greatly secure and help the radiologist performing his gesture. A strong collaboration with the department of interventional oncology from Eugène Marquis Center of Rennes and the imaging and signal laboratory (LTSI) from the University of Rennes I will be needed. This PhD thesis would be a further step towards secure, robust and straightforward planning and assistance of radioembolization procedures.

Phd thesis objectives:

In this context, the PhD candidate will have to

• Perform a state-of-the-art of methodological tools applied to interventional oncology, as well as competitive intelligence searches
• First focus on the preparation and planning of radioembolization procedures using dedicated segmentation and multi-modality registration techniques
• Then use these information intra-operatively to guide and secure the interventional gesture by means of multi-modality registration techniques
• Strongly collaborate with the department of interventional oncology of the Eugène Marquis center in Rennes
• Closely work with the software engineers from the Therenva team to propose a first solution dedicated to planning and guidance of radioembolization procedures

Candidate profile

• MSc (M2), engineer school or equivalent diploma in computer science or electrical engineering
• Skills in image processing, applied mathematics and/or computer vision
• Skills in several programming language (e.g., C++, Python, Matlab)
• Synthesis and editorials skills are required
• Fluency in English (oral and written) is required.
• Recruitment of international English-speaking students accepted

To apply, please send a cover letter and a resume to florent.lalys@therenva.com