3D anatomy project scope – phase 1 – focus on a limb joint

We can enumerate several technical challenges of making high-quality body changes possible. Here, we will start attacking one specific challenge – starting to establish an end-goal for transformation. Specifically, we want to start describing a body’s anatomy, with a reasonable physiological basis, in a flexible way. We ultimately envision a user-friendly interface, that enables immediate feedback on various changes to an anatomical Character. The user should mainly interact at a high level – working with whole limbs, broad tissue types, and changing the shapes of collections of tissues (such as cheeks) without worrying about specific cheek bones, cheek muscles, the cranial nerves, oral cavity, etc. However, all of these details must underpin the Character, and should be accessible in a granular fashion to the user as needed (such as muscle placement, ligaments, bone shape, etc).

In this project, we initially want to develop a small-scale prototype or proof-of-concept, focusing on one anatomical region – the elbow. Using C#, we will develop an object-oriented system handling anatomical data and logical operations, and we will display that data, as well as the consequences of edits to that data, using the Godot engine. While we develop the pipelines and logic underpinning the program, we will gather or generate 3D meshes corresponding to multiple tissue types, such as the bone, muscles, and tendons, around the elbow, and use the program to annotate and precisely edit relevant parameters of those tissues. Once those basics are satisfied, we will focus on further improving correctness and usability, and improving the programmatic representation of anatomical features. Finally, we will use the program to analyze anatomy on a comparative basis. We plan on publishing results, the code, the API, and documentation as we approach completion of this project.

Future directions that will be enabled by this project include: (a) adding more anatomical sites, (b) progressively reducing dependence on vertex/mesh-centric modeling, (c) improving the flexibility of editing, and (d) adding more types of tissues.

Deliverable 1 – Creation of a basic system capable of representing demarcated features of bones, muscles, and other tissues, and editing them in simple ways.
Deliverable 2 – Use of the system to represent anatomical parts and features around the elbow.
Deliverable 3 – Improving usability of the system to allow it to become more widely available.
Deliverable 4 – Detailing of the anatomical system to allow intelligent editing and animation of features.
Deliverable 5 – Demonstration of the system in reasonably representing real anatomical diversity.
Deliverable 6 – Public release of findings and advancements.

This project is still in its early stages. We’ll keep you updated as we get closer to having results we can show.

Check out our GitHub! The code is starting to take shape. Our project has a GPLv2 open-source license.