3D Reconstruction from section images

Hi everyone, I’m quite new to jMonkey and I have just started to work for for a medical company. They need me to implement an application that needs to be able to recontruct a jaw in 3D from the DICOM image files of the jaw’s sections. So, I’m wondering if that kind of application can be done with jMonkey, and I would need to be in real time so we can move the camera and interact with the 3D model. What do you think, would it be possible to start this project with jMonkey?


If you want to reconstruct a 3D object from a dataset (as the DICOM) you probably have to use Marching Cubes algorithms or some Volume Rendering. Seeing this is an engine, it is probably possible but why would you do it this way, have you taken a look at software that was written for this purpose yet? I followed a course on my university in which we used Devide, perhaps you want to look into that, even if you don’t want to use it it might provide you some ideas anyway.

Thanks. I’ll take a look at Devide and see if I can do something from it!

By the way, will I be able to apply interactive functionalities to the 3D scene in devide such as adding elements with the mouse or something?

As far as I understand, it’s a VTK/ITK toolkit that allows you to write your own modules, thus increasing development rates for any tools using medical data. So yeah, you will be able to add elements and whatever you want, you can even add simple shapes using python, it is actually meant to support you doing that. The question is moreover how much time you are going to spend implementing it, as I believe pretty much anything is possible, it just depends on your insights and qualities :wink:

Basic point adding functionalities already exist in the Slice3D module by the way :wink:

I have plenty of time because this is a 6 month project and by the time being I’m spending about 4hours a day in research and implementation.So I should be able to finish it though I’m still quite new to python an never worked with vtk/itk before. Btw, I’m very grateful for your help, you simplified much my work because my first thought on the approach of this project was going through it entirely with C++/OpenGL and so implementing methods like marching cubes and stuff