Current master thesis

M.Sc. Candidate Sara Zucchelli
Supervisors Elena De Momi, Alice Segato
Title 3D Motion Modeling and Control of a Steerable Needle for Keyhole Neurosurgery
Collaborations Imperial College London
Description StefanoNigrisThesis

Nowadays minimally invasive brain surgery is a growing field due to the exponentially increase of cerebral diseases. EDEN2020 European project proposes a change in the treatment of brain diseases, thanks to the integration of five different technologies for minimally invasive brain surgery. This thesis it is conducted in the framework of this project, in particular it is focused on the low level control of a medical robot device that will move the EDEN2020 catheter, a multi-segment probe made out of four interlocked segments, that can slide along one another generating curvilinear trajectories. The aim is to develop a control model of the catheter considering its cinematic constraints, including the maximum curvature. The dynamic model will take as input the desired trajectories in output of the EDEN2020 simulator, by using ROS (i.e. Robot Operating System). Moreover, it will be applicated to the model of the brain the consistence characteristic of the real brain to make the simulation more realistic.

M.Sc. Candidate Valentina Corbetta
Supervisors Elena De Momi, Alice Segato, Francesco Calimeri
Title Answer Set Programming for Steerable Needle Path Planning
Description StefanoNigrisThesis

Last-generation steerable needles represent a breakthrough in keyhole neurosurgery, as they are capable of reaching targets behind sensitive structures or in previously impenetrable areas. Different path planning approaches have been developed and can be divided in classical methods and machine learning. The first ones require a tradeoff between completeness and efficiency, which often makes them unsuitable for complex search problems; the latter requires large datasets to train the models, such datasets can be hard to retrieve in the medical and surgical context. The aim of this thesis, which is part of the EDEN2020 project, is to develop an artificial intelligent agent programmed using Answer Set Programming (ASP) to perform path planning of a steerable needle catheter to perform Deep Brain Stimulation . ASP is a form of declarative programming used to solve difficult (especially NP-hard) search problems. It constitutes a valid alternative to the previously introduced approaches, as programs are very compact and fast and it does not need large datasets. The agent should be able to autonomously compute the path from the starting point to the target, avoiding obstacles and satisfying constraints which can be customized by the surgeons and clinicians based on the specific case. It should also be able to classify a list of previously computed trajectories, based on their optimization of the specified constraints. The agent can be used both in pre-operative and intra-operative settings.

 

This site uses cookies to enhance your experience. By continuing to the site you accept their use. More info in our cookies policy.     ACCEPT