3D modeling of the nerves of the left upper limb - Discussion

This research work on the nerves of the left upper limb and on the roots of the brachial plexus on the left is part of a larger project led by Professor UHL, which consists in digitally modeling the entire human body in 3D.

From our first course of the University Diploma of Clinical and Digital Anatomy of the University of PARIS DESCARTES, each student has chosen a subject corresponding to a part of the anatomy of the human body.

This one naturally presenting structures as diverse as varied, the list of the subjects was very diversified: right kidney, veins of the right upper limb, ligaments of the upper limb, splenic arteries, pulmonary veins, etc.

I was thus able to choose to model the nerves of the left upper limb in men and the roots of the left brachial plexus in men.

Once the subjects were distributed, a few hours of class were necessary to present the different functions of the computer modeling software to be used.

Several hours of personal use completed the training in order to learn how to concretely master this "Winsurf" software.

The first step of this project of modeling the nerves of the left upper limb and the roots of the left brachial plexus did not pose any particular problem. It consisted in a bibliographical research in order to master the theoretical knowledge of the approached structures. A research work was thus carried out in order to master the theoretical descriptive anatomy of the nerves of the upper limb and the brachial plexus, in particular their topography with respect to the surrounding structures but also their collaterals and terminations.

To work, the Winsurf software is necessary but it is only functional from a PC with a non recent Windows version. It is thus necessary to be able to obtain a PC or not to be able to use the Winsurf software. In addition, backups of the models had to be made on a regular basis because the program was regularly shut down and all unsaved data was lost.

This second step proved to be extremely tedious since it consisted in recognizing the various entities on the anatomical sections provided. It was thus a question of concretely identifying the nerve sections among the other structures of the left upper limb, thus passing from the theoretical to the concrete.

Several obstacles were encountered at this stage of processing the subject, as the recognition of the nerves of the left upper limb was a painstaking and difficult task in several respects.

Their small size makes them very complex to identify. The "Winsurf" software adds a technical difficulty since it only allows zooming in on the sections up to a certain point, the image deteriorating beyond that point due to pixelation.

Finally, some nerves did not appear clearly on the sections because of their horizontality. This was notably the case for the axillary nerve. It was then necessary to guess their presence and their course with the help of theoretical knowledge and adjacent structures.

Due to the pixelation of the sections, the location of the musculocutaneous nerve was impossible when dividing it into its terminal branch. Even with guessing, the tracing was not coherent and I had to choose not to do it rather than invent it.

Similarly, the brachial plexus could not be modeled, only its roots and endings. The pixelation of the sections as well as the horizontality of certain structures of the plexus made this work impossible.

Thirdly, the realization of the "contouring" itself could start. It was necessary to follow the contours of the entities identified with the cursor proposed on the Windsurf software, while managing to find a balance between anatomical and aesthetic reality.

This step of realization of the 3D modeling presented two main difficulties, both related to the final aesthetic rendering.

The first difficulty was to determine on which distance between the cuts it was necessary to work. A work on a rectilinear entity with a large volume can be done every 20 cuts. On the contrary, a work on a curved entity with a small dimension will require a more precise work and thus to take closer cuts.

Since the work on the nerves of the left upper limb falls into this second category, we first had to look at all the slices, i.e. with an interval of 1 slice, and then contour on an interval of between 5 and 20 slices.

However, the closer the work is done with cuts, the more the aesthetic result decreases since the result does not appear smooth but with a jerky effect. It is therefore necessary to gauge the maximum distance between the cuts that can be used without missing certain details and curves of the entity to be outlined.

The second difficulty encountered in the "contouring" stage was to know how to place the points from one cut to another. At first sight, this work seems simple and only requires the use of the software tools. However, it quickly becomes clear that a correction must be made in order to avoid a jerky layout. Thus, it is important that from one cut to the next, the point selected is as close as possible in X and Y to the previous cut and the next cut in order to obtain a harmonious entity. The result obtained is thus smoother and more aesthetic, although further from the anatomical reality. Here again, it is a matter of finding the right balance.

Marie Messager
Osteopath in Versailles
78 - Yvelines