What’s involved in developing anatomical models from the ground up? We are very proud of our models and we would like to share the process we go through in developing them. Whereas our previous models were developed for rendering images, our new models have been completely rebuilt from scratch specifically for our 3D Display engine to improve functionality, medical accuracy and performance.
At 3D4Medical we pride ourselves on staying at the cutting edge of technology. To this end we are introducing a number of new functionalities to our 3D apps. A slices / scan feature, that will allow the user to pull a cross-sectional plane through any of the models onscreen to obtain a cross section view. A muscle animation feature, which will allow the onscreen model to be animated, illustrating muscle movements.
New demands are being placed on our apps that require a greater level of anatomical detail and accuracy. Greater detail means larger models and more pressure on processing power, so new modeling methods were needed to accommodate this greater level of detail.
To date we have developed solely for the Apple app store. Our challenge was to be able to transpose our 3D Display technology onto both Android and Windows 8 platforms. To do this, processing power was a real concern. New custom-built models were the key to achieving the necessary increased performance.
A wide variety of reference materials are used including a real human skeleton. Measurements and multiple images are taken from every angle.
Normally the entire modeling process would take place in a 3D software package. However our modelers need to work in close conjunction with our proprietary 3D real time engine. All models are built in conjunction with our 3D Engine as every step has to be scrutinized to ensure maximum compatibility.
Once the basic modeling is completed, it is exported to another package where finer detail is sculpted into the model. This results in what’s called a ‘normal map’ of the model.
The normal map created is then brought into another 3D software package and ‘unfolded’ so that a texture can be applied. From this point, a texture artist works with the new model. Using reference photos, areas are cut out and merged together in the software package.
When completed, normal maps from the two software packages are used in conjunction with each other to add even more detail to the model. The completed texture is then applied to the model.
The model is then imported into the 3D4Medical proprietary development tool where the edge seams are blended out. It is in the 3D4Medical proprietary development tool that all of the associated media is assigned, such as audio, hotspots, text and behavioral instructions.
In addition to the way the new models were constructed to ensure a low polycount, they also undergo a final optimization for our 3D engine. This is to save memory for whatever platform it is developed for.
The model is finally imported into our proprietary real time 3D display technology to create our trademark immersive learning experience.