InitRech 2015/2016, sujet 8

Summary

With the aim of training new surgeon-students, HAL proposes a computer-based training system for cataract surgery in this paper. Indeed, this operation, to be well done, needs a lot of practice. However, sutends can't train themselves with real patients. Moreover, this training system allows students to make some mistakes and keep training, which is not the case in reality.

The system described is thought to be as realistic as possible, for instance there is a 3D mannequin to realize operations, a full environment which looks like a true surgery room (with monitors, surgery tools, ...). This training system is based on a software: SOFA. In this, all is calculated and close to the reality: real values are used to configure the tools, the deformation of the capsule or the lens isn't random but genuine.

Firstly, the deformation of the lens is represented thanks to a co-rotational FEM (Finite element models) formulation with linear elasticity and tetrahedral mesh tessellated in order to produce small elements. With this method, all the calculations aren't done so it dosen't need a lot of time to change the topology of this mesh, that's why it's interessant. Moreover, the surgeon-student can use the phacoemulsifier in two different ways: Aspiration or Fragmentation. The aspiration is used to move liquids from the eye to the tip whereas fragmentation breaks the lens due to the tip's vibrations. During an operation, the quantity of light that is bouncing back from the retina through the lens is a really important signal for the surgeon, this effect can be reproduced thanks to 'Projected Tetrahedra'.

Secondly, they focused on Continuous curvilinear capsulorhexis, which its main aim is to create a perfect opening avoiding some tear during the surgery.To model the deformation and tearing of thin soft tissue, a transversely isotropic elastic model is used. Thanks to this model, the direction of tear's propragation can be known. In order to reproduce the real « tearing » a work has been done about the direction of fibers within the soft tissue. To do that, a transversely isotropic FEM formulation is used. Another part of the work has been done on determination of the fracture direction and the fact to restrict the potential fracture by two solutions. Thanks to all this work, it's possible to create realistic deformations and a good quality's tearing.

Thirdly, they focused on intra-ocular lens deformation. To simulate this, a superimposing with a plane stress and bending energies is the solution choosen and that offers a good rate between computational efficiency and accuracy.

Then, they focused on the interaction devices. Indeed, this system needs to be robust because surgeons and hospitals will be the targets of this project. Thei approach is based on infra-red optical tracking, actually the system includes 6 infra-red cameras in order to cover all the area and track the instruments. This job of tracking is also realized thanks to the software Tracking Tools which uses markers on rigid bodies. To make dialog the two computers, the VRPN library is used. One of the two computers is the tracking-computer, whereas the other has a better graphic card to run the simulation.

Main contribution

This article's main aim is to introduce us a Computer-based training system in the field of surgery. More precisely, the main contribution of this project is to help new surgeons for the cataract surgery. This project is really usefull and interesting because the students can train themselves with a very realistic environment and a 3D mannequin. The cataract surgery is more difficult than other some operations and thanks to this project, the new profesionnals can make some mistakes, fail, but they can do this simulation again and again in order to improve their skills. We all know that a simulation isn't the reality and that all hazards can't be reproduced with this training but it can be usefull for the student, in my opinion, to practice himself before doing a real operation with real risks at the first error. In this project, everything is taken into account in order to make this system of training as close of the reality as possible, that's why after exercising with this simulator, the students will be able to do a real surgery without significant changes except exceptional situations.

Applications

This computer-based training system could have many applications in the future. Of course, the main application, which we all think about, is to give students the opportunity to try it. For example, in medicine shcools students in surgery will improve all their skills with this system, during the studies so they'll be able to do correct operations as soon as they'll be graduated. Actually, some written classes could be replaced by this system of training.

Another application for example, is to offer to high school students to try this system, so they can see what is catarct surgery and let them know if they would work in this field before the post degree's choice.

As we know, a lot of people who will be operated fear before the D-day, maybe this could be reduced if they see what is the cataract surgery, how the surgeon do the operation, how many time it takes, how is the environment during the patient is sleeping. Indeed I think to that propose to the patient to see a whole operation could reassure him.

Moreover, this project can be usefull for many hospitals because surgeons can err without any risks, so if a mistake recurrent in the cataract surgey, they could be think together about a solution to this issue. In the long term, this system could be a reference for all the cataract surgeons, that means all the cataract's operations must be realized ine the same way, all over the world. This potential application could allow a surgeon to know what was wrongly done by a colleague during the surgery even if he wasn't in the same room this day.

Further to the previous application pronounced, engineers could create a a robot which can realize this cataract surgery. In fact, if there is a standardization of this operation, a robot can be able to do this. Obviously, it would make disappear the human-side adn the contact with the patient, but hospitals will save a lot of time and surgeons could be available for other actions which can't be standardized. Moreover this idea of a robot could erase all notions of risks for the patient.

We also can imagine that, if the results of many tests are good with this system, it can inspire other research's groups in order to build simulation systems. Maybe in other particular surgery, or in a totally different field. Even if it already exists simulation applications (in the aeronautic field for instance), a lot of job could be better done with simulations before.