Substrate shapes generation for additive manufacturing of medical corsets tailored to various patient morphologies

Abstract

Additive manufacturing (AM) recently appeared as an alternative to thermoforming for medical corset production, enabling lighter, more breathable, and highly customizable designs by creating complex shapes and adjusting material properties. However, the common 3-axis architecture of AM machines limits corset performance due to the unidirectional layering of material. As an alternative, a method involving radial material deposition on a rotating substrate is being explored. In this setup, the substrate’s shape becomes a key factor in reducing manufacturing time and cost, directly impacting the needed volume support between the substrate and the corset. In order to avoid the time and material-intensive practice of creating a new substrate for each corset, the objective is to use a limited number of reusable substrates specific to morphological categories of corsets. A four-step method for generating optimally shaped substrates is proposed. Firstly, unsupervised classification of 3D patient scans sorts them into fixed morphological categories. After achieving the optimal 3D alignment of the scans per category, substrates can be generated from the resulting minimal surfaces. A validation process involving fitting each scan to each categorized substrate and then calculating the resulting intervolume demonstrates the potential to reduce support volume in the additive manufacturing of a morphological class of corsets.