Skip to Main content Skip to Navigation

Développement, optimisation, évaluation d'un système de réalité augmentée en chirurgie laparoscopique

Abstract : We developed an Augmented reality (AR) system that can guide the surgeon during laparoscopy. Augmented Reality (AR) is a technology that can allow a surgeon to see subsurface structures. This works by overlaying information from another modality, such as MRI and fusing it in real time with the endoscopic images. The surgeon can easily localize tumor in the parenchyma of an organ. He could also easily localize all anatomical and structural landmarks available on the preoperative imaging. Three phases are necessary. Firstly, Preoperative MRI data are used (segmentation) to construct 3D mesh models (external surface of the organ, tumor(s), anatomical landmarks). During the surgery a 3D mesh model of the organ is constructed and an initial registration phase (using a deformable model of the organ) is performed. Then the third phase is a real-time tracking phase. AR has never been developed for a very mobile organ like the uterus and has never been performed for gynecology. Our system works for soft and mobile organs. We used two tumor models: myomas in gynecology and kidney’s tumor for partial nephrectomy. For each type of tumor we used the same step to develop our system. First step: experimental tumor model, second step: ex vivo improvement and first clinical evaluation, third step: clinical evaluation. We are in the third phase for myomectomy and in the second phase for partial nephrectomy. In our uterine model (3D printed uterus), AR improves localization accuracy of the myomas compared to the classical localization method (MRI only). This was the first user study to quantitatively evaluate an AR system for improving a surgical task. After optimizing our system using ex-vivo data, we tested it during laparoscopic myomectomy and demonstrated the feasibility of the real-time tracking and localization of the myomas and of the uterine cavity. For partial nephrectomy we created a kidney’s tumor model (porcine model). Our study shows that AR allows accurate localization of very small tumors and improved the mean accuracy of tumor resection, with higher rate of free margins around the tumor. We are currently developing phase 2. For liver resection the development is in phase 1. Our team is a mix of engineers, scientists, doctors and surgeons. AR is a very promising technique with large applications. It allows displaying all preoperative imaging data on our laparoscopic screen. AR should be used soon for other gynecological, urological and digestive pathologies. Other technologies (including deep learning) should allow major improvement of our system.
Document type :
Complete list of metadatas

Cited literature [172 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Wednesday, July 1, 2020 - 4:45:32 PM
Last modification on : Friday, October 23, 2020 - 5:04:18 PM
Long-term archiving on: : Wednesday, September 23, 2020 - 5:08:00 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02886659, version 1



Nicolas Bourdel. Développement, optimisation, évaluation d'un système de réalité augmentée en chirurgie laparoscopique. Médecine humaine et pathologie. Université Clermont Auvergne, 2017. Français. ⟨NNT : 2017CLFAS021⟩. ⟨tel-02886659⟩



Record views


Files downloads