Algorithmic detection and categorization of partially attached particles in AM structures: a non-destructive method for the certification of lattice implants

What is it about?

Additive manufacturing (AM) is a process in which structures are built layer upon layer using an input material; for a metal powder input, structures are built by fusing powder together using an energy beam within a powder bed. Due to the build mechanism, partially melted powder residue may remain within or attached to manufactured structures. As AM can be applied to different fields, it can be imperative, such as for dense, lattice based, orthopaedic implants, to verify that the residual powder be removed. Currently, there exists no formal methods to detect and categorize this residue. To overcome this, a novel method is proposed in the form of a software algorithm. The algorithm takes in computed tomography (CT) image data and analyses it. After algorithmic analysis, a visualization can be generated depicting the scanned structure and partially attached particles, and histograms are generated based on characterizations of the different properties of the particles. We propose this methodology provides a formal basis for the design and certification of powder-based AM implants.

Featured Image