InitRech 2015/2016, sujet 14

Content

Sujet: Local implicit modeling of blood vessel for interactive simulation

SUMMARY

The article argues mainly about the interest of computer simulator in the interventional radiology.It is not an easy task to realize or to create a real-time simulator.The main problem is to find solution when a blending occurs between 2 vessels(eg.:internal hemorrhage,aneurysm...). Some engineers focused their interest on the real-time simulation in terms of vizualisation by proposing an algorithm model. The algorithm model embraces 5 points.The first one consist into representing a surface by a mathematical formulation.With this formulation,they have been able to do a sum of spheres. After the conceptual phase,they deal with an other point:minimization of the energy in order to avoid collision.As we know,every body's cell needs energy to be functionnal.A surface can be describe as a plan,which means that it is defined by many points(cells).The Lennard-Jones energy validates the model. They also tried to subdivise the blob whose isosurface is closest the Pi;the point which is farthest to the surface.the subdivision is operated in large areas.

They consider an optimization subdivision.The subdivision can occasion a high number of blob sections.Which is not reliable,because may lead to a collision. They use the model for simulation by taking due account the equipment used in medical field.

They also made experiences by using 5 patient data.Data involve the internal carotid artery.The Cauchy algorithm has been used for its efficiency and Lennard-Jones energy(described above). However,some geometric errors occur and had to be fixed by Taubin's approximate distance.But 2 parameters had to be taken into account such as the distance between 2 blobs and the number of subdivision.

The other point to tackle was to quantify the efficiency of the model and computation time.3-D modelizing one patient take an half hour.The purpose was to compare the number of blobs and the number of triangle on the isosurface.In one hand,the test was conclusive but in other hand,the model was not able to simulate the vessel due to discontinuities. In conclusion, the model has been integrated in the simulation software and tested on a specific patient.In the future a global perspective within sight is to use make simulation with image data.

MAIN CONTRIBUTION

First of all,the main contribution is that they had been able to simulate differents shapes and states of vessels in real-time.Secondly,this is very important for treating aneurysm (which is an abnormal swelling of vessels)in order to prevent the rupture. This concerns not only aneurysm but also others vascular diseases.

An other aspect is that the effect of blood flow can be analysed and understood.This can be an indicator of vessels'dysfunction. Some flexible medical devices had been also simulated such as a catheter or a guide.

SOME APPLICATIONS

There are a lot of applications especially in the medical domain.Those kind of researches will help to resolve important issues in human arteries in vascular disease characterization and assessment.The algorithm model combined with the 3-D modeling will provide a suitable modelisation of blood vessels. With recent advances in medical imaging,it will be possible to treat and analyse many diseases with anatomic data specific to the patient.

As an example,during the intervention,it is better for a doctor to visualize exactly what is happening in real time.For example,in case of internal hemorrhage,cerebrovascular accident,or liver treatment,goitre and thyroid dysfunction... With technological advance,precision is guaranted and a reduction of chirurgical accidents. Perhaps,it is also conceivable the treatment of obesity;a disease mainly caused by a high level of LDL(Bad Cholesterol,in this context,LDL is responsible of atherogenesis).This modelisation will help to move around the vessel and localize the lesion area.