Model and Image Controlled Prostatic Robot – PROSBOT

This project addresses the topic of computer assisted medical interventions (CAMI) and its medical domain is prostate cancer (PCa).
Prostate cancer is the most frequent cancer for men with 60,000 new cases every year in France and 9,000 deaths. Its incidence has been constantly growing in the last 20 years, due to an increase in life expectancy, but also due to the diffusion of opportunistic screening based on the evaluation of the prostate marker PSA.

The European Randomized Study of Screening for Prostate Cancer (ERSPC), performed with 182,000 men, showed that systematic screening allowed to reduce the specific death rate by 20%. However, it also highlighted that to “save” one patient, 1500 had to undergo screening and 50 had to undergo surgery. This is problematic when considering the fact that prostate cancer treatment has an important cost for the society and can induce complications like incontinence and impotence, which notably alter the quality of life of the patients.

Two questions can be identified in this context: 1/ how can we improve diagnosis so as to differentiate patients who need a treatment from those who need only active surveillance; 2/ how to make radical treatments evolve to interventions focalized on the cancer.

In that context our project will focus on prostate biopsy intervention. It is the sole intervention allowing to confirm the diagnosis, to evaluate the prognosis of the disease and even to select and to plan the treatment. The tissue samples are acquired under ultrasound control and through rectal access. This gesture requires dexterity to reach targets with precision and seems barely reproducible between different operators. Currently clinically evaluated commercial products for computer assistance exist; they are systems for navigation or needle positioning (in particular robotic systems), either through transrectal or by transperineal access. Whatever approach, they still do not handle constraints concerning the mobility and the deformation of the prostate, and neither the ergonomics of the gesture. The challenge in this domain is hence to invent and to associate complementary technologies in a system that answers all these constraints while flawlessly integrating into the clinical and regulatory environment.

We propose a unique co-manipulated and image-guided robotic device, not to make an earlier or more frequent diagnosis of prostate cancer, but to improve the precision of the gesture with the objective to improve the quality of the diagnosis and to improve the decision-making for therapy. It will be integrated into the product UroStation® from KOELIS company, which is equipped with an innovative biopsy mapping system, for an integral management of the quality of the intervention. The device will be equipped with a biopsy simulator for efficient learning of the gesture and the handling of the device. The system will exploit advanced algorithms for ultrasound image analysis and processing, real-time deformable biomedical models of the prostate and a robot controller for image and model based co-manipulation.

This industrial research project will be implemented by the following partners:
(1) The company KOELIS, coordinator, CAMI specialist in the urology domain,
(2) the laboratory TIMC-UJF in Grenoble, expert in the domains of biomedical models, processing and registration of images, and robotics,
(3) the institute UPMC-ISIR, specialized in robotics and co-manipulation,
(4) the department of urology of the APHP Pitié Salpétrière in Paris, which made the clinical demonstration of previous systems with KOELIS, TIMC and ISIR and
(5) the CIC-IT of Grenoble, a centre for clinical investigation of technological innovations, expert in the formalisation and evaluation of the expected and effective return of new devices.