In vitro method to quantify and visualize mechanical wear in human meniscus subjected to joint loading

Guidelines for Accurate Multi-Temporal Model Registration of 3D Scanned Objects

“…These pilot tests also showed that increasing the block size above 4 cm 2 had a negligible effect on the registration error. A replica of a human tibia bone was selected as the object of interest [ 19 ]. Within modeling software (Meshmixer, Autodesk, v3.5), three separate 3D models of the tibia bone geometry and the 3D registration blocks were imported as object files.…”

“…The 3D scanner was used to generate two identical 3D models to serve as the reference and target following the experimental method ( Figure 3 ). All scanning was performed in a dark room, with the only light source coming from the projector [ 10 , 19 ]. This setting was used to minimize the presence of ambient light that could impact scanning quality.…”

“…The projector field of view (FOV) was 200 cm 2 (resolution = 0.12 mm). The exposure settings were set to high dynamic range (HDR), which automatically determined the optimal exposure (range of exposure times = 16.67–150 ms) [ 19 ]. The tibia was scanned every 30 degrees using the rotary table ( Figure 2 ) for a total of 12 scans, resulting in a full 360-degree captured view.…”

“…Many studies have used this approach for multi-temporal 3D model registration by selecting a wide variety of features. These features have consisted of either intrinsic features (natural to the object) [ 8 , 9 , 11 , 17 ] or artificial features (intentionally placed marker) [ 4 , 18 , 19 ]. For example, Valigi et al evaluated wear in an artificial knee by selecting unworn portions of a tibial component to register knee models that were generated at different time points (i.e., pre-wear and post-wear) [ 11 ].…”

“…For example, Valigi et al evaluated wear in an artificial knee by selecting unworn portions of a tibial component to register knee models that were generated at different time points (i.e., pre-wear and post-wear) [ 11 ]. More recently, we measured in vitro fibrocartilage wear in the human knee by selecting artificial features fastened to the tibia bone to register 3D models before and after joint loading to simulate pre- and post-wear time points [ 19 ]. The accurate application of this common registration method is dependent on the local artificial or intrinsic features selected by the operator to calculate the registration transform, yet there are limited guidelines for selecting these features.…”