An application of the scaled–pixel–counting protocol to quantify the radiological features of anatomical structures of the normal tarsal joint in horses

Background: As high-performance human and equine athletes train and compete at the highest level of effort, the prevention of high-performance-cased diseases, such as osteoarthritis (OA), requires knowledge of the anatomy and physiology of the subjected bones. Objective: Implementation of the scaled–pixel–counting protocol to quantify the radiological features of anatomical structures of the normal equine tarsal joint as the first step in the prevention of the tarsal joints OA in high-performance sport horses. Methods: A radiographic examination was performed in six cadaverous equine pelvic limbs. The dorso–plantar projection of the tarsal joint was performed using density standard (DS) attached to the radiographic cassette, standard X-ray equipment and standard diagnostic imaging protocol. On each of the radiographs, pixel brightness (PB) was extracted for each of the 10 steps (S1–S10) of DS. On each of the radiographs, seven regions of interest (ROIs) were annotated representing four bones (II tarsal bone [II TB], III tarsal bone [III TB], IV tarsal bone [IV TB] and central tarsal bone [CTB]) and three joints (proximal intertarsal joint [PIJ], distal intertarsal joint [DIJ] and tarsometatarsal joint [TMJ]), respectively. For each ROI, the percentage (%) of number of pixels (NP) from each range was calculated. Results: The % of NP was lower in bones than in joint spaces for S1–S6 and was higher in bones than in joint spaces for S8–S10. The % of NP was higher in PIJ than TMJ for S1 and higher in PIJ than DIJ for S4. No differences were found between consecutive bones for all examined steps of DS. Conclusions: An application of the scaled–pixel–counting protocol provides the quantitative radiological features of normal bone and joint structures of the tarsal joint in horses, making possible differentiation of the lucency of joint space and opacity of bone structure.