More than 100,000 anterior cruciate ligament reconstructions are performed annually in the USA. The hamstrings and the patellar tendons are the most frequently used graft tissues. Up to ten percent of these grafts are deemed to have failed, generating considerable discussion in the literature regarding the ideal graft choice. A three-dimensional computational model, taking into account both material and geometrical non-linearities, would be useful in predicting the performance of different graft tissues and fixations. Unfortunately, the mechanical characteristics and parameters needed for such a model are complex and largely unknown. The aim of this study is to develop a method for measuring the geometrical properties needed as input for a three-dimensional tendon model.

A laser-based, non-contact technique is used to generate a series of cross-sectional profiles along the length of the tendon. Unlike previously proposed methods, it is able to detect concavities and can be constructed using equipment commonly found in an engineering laboratory. A laser line generator (Stocker-Yale Lasiris SNF, Quebec, Canada) projects a horizontal line onto the sample. Images of the line are acquired with a digital video camera (Basler A631fc, Germany) as the tendon is rotated. These images are reassembled into 2-D slices using MatLab software. Multiple cross-sections can be combined to create three dimensional geometries.

The new method was validated on objects of known shape (circular and hexagonal cylinders). The cross-sectional area measurement was found to be accurate to within 2.5%. The method was repeatable to within 1.7%. Six bovine flexor tendons have been analysed; concavities were evident in four of these. This method could be adapted to determine the surface geometries of other long and slender objects.