2024-2027

ThermoNav - Automatic Multi-Needle Adaptive Precision in Percutaneous Thermal Ablation Planning

 

Thermal ablation therapies are particularly complex procedures, especially when they involve
multiple needles. They result in high local recurrence rates, ranging from 10% to 39.1% (within 5 years), due to insufficient coverage. In this PhD, we aim to provide a comprehensive solution to the problem of multi-needle planning, intra-operative guidance, and user-friendly replanning, with a focus on simulating and visualizing thermal propagation and predicting cell death.

  • PhD student: Jonas Mehtali
  • Supervisors: C. Essert, J. Verde

Geometric Weakly Supervised Approach for Multimodal Alzheimer disease detection and Monitoring : translational study in mice and human

 

Previous research in medicine demonstrated that Alzheimer disease (AD) affects Hippocampus shape and the gait. In this thesis, we aim to investigate a joint analysis of 3D hippocampal shape and skeleton-based gait for AD diagnosis and monitoring. We propose to focus our first experiments on transgenic mice then we aim to transfer the retained results on human data. Moreover, we will target the exploration of the active learning techniques to reduce the human manual labelling effort.

  • PhD student: Mubarak Olaoluwa
  • Supervisors: H. Drira, L. Harsan

Uncertainty quantification in automatic delineation for radiation therapy

Recent advances in automatic segmentation using deep learning have led to the emergence of
solutions for the delineation of structures in radiotherapy (tumors and organs at risk). These
solutions have the potential to improve patient care by saving time and reducing variability. The goal of this thesis is to study the uncertainty associated with the predictions of these algorithms.

  • PhD student: Tristan Kirscher
  • Supervisors: P. Meyer, S. Faisan

Combined Raman spectroscopy and full-field optical coherence tomography applied in oncology

Correct assessment of tumor margins during surgery is an essential prognostic factor in avoiding re-operation and the development of metastases. The aim of this thesis is to couple two imaging techniques, Raman spectroscopy and full-field OCT (FFOCT), to offer surgeons a new tool for rapid, quantitative evaluation of tumor margins during surgery as no adapted parallel intraoperative technology exists.

  • PhD student: Phu Duong Le
  • Supervisors: H. Salehi, A. Venkatasamy

INTERACT: Investigating iNTraoperative Events to Reveal Anastomotic Complications in colorecTal surgery

The project aims to enhance anastomotic leak (AL) prediction models by integrating intraoperative data alongside preoperative factors. Focused on tissue perfusion, particularly through indocyanine green fluorescence angiography, we seek to identify perfusion patterns correlating with postoperative complications, notably AL. Leveraging the dataset generated in the CompSURG project, we will explore transfer learning techniques to extract perfusion information from laparoscopic and endoscopic videos.

  • PhD student: L. Arboit
  • Supervisors: N. Padoy (IHU-ICube) & G. Quero (surgeon, digestive surgery unit, Fondazione Policlinico Universitario Agnostino Gemelli IRCCS, Rome)