PhD: Endovascular Navigation Haptic Training

Student: Kenza OUSSALAH, web page of LaMCoS

Supervised with Benyebka Bou-Saïd from LaMCoS research lab and Richard Moreau

Period: started in 2023, to be defended in 2026

Research project: ANR RHU IceLand

Financed by ANR

Keywords: Haptics, Simulation, Hands-on Training, Endovascular navigation

Summary

Context

This PhD is part of the ANR RHU ICELAND project (ANR-21-RHUS-0006-04), which aims to develop a learning simulator for endovascular surgery on the heart’s mitral valve. This simulator can also be used prior to surgery, using patient imaging to enable the surgeon to validate their surgical strategy (choice of tools and trajectories). If the mitral valve is damaged or misaligned, it cannot close completely with each heartbeat. Some of the blood flows backward with each contraction. This reflux, called mitral regurgitation, puts pressure on the heart and can cause symptoms such as shortness of breath and fatigue, leading to atrial fibrillation or heart failure. It is one of the most serious cardiovascular diseases. Cardiovascular disease is the leading cause of death in the industrialized world. These cardiovascular diseases are expected to increase with the aging of the population and obesity-related problems.

Among common vascular diseases, the treatment of mitral valve regurgitation is an important part of cardiovascular surgery. The total cost of cardiovascular disease in Europe is estimated at around €196 billion. For several years, minimally invasive surgery has been a common alternative to open surgery, reducing surgical risks, hospital stays, and blood loss. However, the success of this type of operation depends largely on the surgeon’s expertise, which is difficult to transfer to novices. Novices learn the technical procedures by observing experts and through mentoring. However, this type of training is not ideal and does not meet the expectations of the HAS (Haute Autorité en Santé), which recommends training on a simulator

Objective

The objective of this PhD is to design and control a simulator for learning this procedure. The simulator will consist of two parts. The first will be a digital model of the vessels from the introduction point to the valve. Deformation during the insertion of the tools will be taken into account, as well as blood flow, which can disrupt the trajectory of the tools. The second part will consist of a dedicated haptic interface allowing the user to navigate the digital model of the vessels and heart while feeling the contacts as in a real surgical procedure. The main contribution of this simulator lies in the coupling between a dedicated haptic interface and advanced biomechanical models. The digital part will be developed by LaMCoS and will reproduce not only the geometry of the tissues but also their mechanical properties.

The goal is to find the right balance between computing time and realism. The haptic part will be developed at the Ampère laboratory. It will need to be able to communicate with the digital part in order to exchange positions and forces when the user performs the action. It will need to reproduce as faithfully as possible the interactions between the tools and the tissues in order to allow for the most complete kinesthetic immersion possible.

Scientific locks

A first scientific challenge is to propose a control law that allows for the reproduction of realistic and relevant force feedback. The question is to determine which criteria are crucial for reproducing adequate contact: how much precision is needed in the force feedback? It will also be necessary to link the parameters of the control law with the mechanical properties of the tissues in order to reproduce their behavior during the execution of the movement. A second scientific challenge is to reduce the calculation time during numerical simulation not only by developing efficient calculation methods but also by optimizing algorithms to converge towards real time. As it may not be possible to obtain a frequency equivalent to that of haptic controls, a third scientific challenge is to compensate in the control laws developed for the differential between numerical calculation and haptic control. In fact, this type of coupling is usually managed by the numerical model. In this PhD, this coupling will be managed by the control laws of the haptic interface to relieve the computer performing the model updates.