Abstract
Measuring articular cartilage thickness from 3D models developed from laser scans has the potential to offer high accuracy. However, this potential has not been fulfilled, since generating these models requires that the cartilage be removed, and previous methods of removal have led to systematic errors (i.e., bias) due to changes in the overall dimensions of the underlying bone. The objectives were to present a new method for removing articular cartilage, quantify the bias error, and demonstrate the method on the distal (i.e., 0° flexion) and posterior (i.e., 90° flexion) articular surfaces of example human femurs. The method consisted of creating a 3D articular cartilage model from high-accuracy (i.e., precision = 0.087 mm) laser scans before and after cartilage removal using dermestid beetles to remove the cartilage. Fiducial markers were used to minimize errors in registering surfaces generated from the two laser scans. To demonstrate the method, the cartilage thickness was computed in distal and posterior subregions of each femoral condyle for three example cadaveric specimens. The use of dermestid beetles did not introduce measurable bias, and the previously reported precision achieved in 3D cartilage models with the laser scanner was 0.13 mm. For the different subregions, the cartilage thickness ranged from 1.5 mm to 2.0 mm. A method of imaging by means of laser scanning, cartilage removal by means of dermestid beetles, and 3D model registration by means of fiducial markers ensured that cartilage thickness on the articular surface of the long bones of the knee was determined with negligible bias and a precision of 0.13 mm. With this method, the potential to measure cartilage thickness with high accuracy based on 3D models developed from laser scans can be fully realized.