Jésica Zuchuat1, Adriana Manzano, Oscar Decco. 1. Bioimplants Laboratory, Faculty of Engineering-National University of Entre Rios, Oro Verde-Entre Rios, Argentina.
Abstract
OBJECTIVE: To analyse a set of morphological features from digital radiographs (RX) and bone mineral density (BMD) values estimated from quantitative computerized tomography scans (QCT scans) of the knee joint of an osteoporotic rabbits, and to determine the relationship and correlation of these variables to be considered as an alternative diagnosis method. METHODS: The knee joint of rabbits (N = 9 ovariectomized and injected with of methylprednisolone sodium succinate (OVX + MPSS) and 3 sham operated healthy controls) were subjected to radiographic examination before the beginning of the study and after 6 weeks; after sacrifice, they were immediately scanned with a 64-channel CT. A set of morphological features was extracted from RX images and then subjected to Principal Component Analysis (PCA); BMD-values were calculated at different depths from the articular surface of the femur and the tibia. The selected morphological variables and the BMD values were correlated in order to determine useful information in medical diagnostics. RESULTS: Ten morphological variables explained 80.39% of the total variability, but only some of them demonstrated significant differences between baseline and OVX + SSMP states and between OVX + SSMP and control. Spearman analysis showed higher positive and negative correlation of these parameters among them (r = 0.67, 0.81, 0.83 and 1); and a moderate correlation with the BMD values (r = 0.50, 0.52, 0.52 and 0.55) during the progression of osteoporosis (OP). CONCLUSION: The Spearman's rank test supported the overall significant correlation between both, the morphological features and BMD values, making them as a reliable alternative option to the diagnosis of osteoporosis.
OBJECTIVE: To analyse a set of morphological features from digital radiographs (RX) and bone mineral density (BMD) values estimated from quantitative computerized tomography scans (QCT scans) of the knee joint of an osteoporotic rabbits, and to determine the relationship and correlation of these variables to be considered as an alternative diagnosis method. METHODS: The knee joint of rabbits (N = 9 ovariectomized and injected with of methylprednisolone sodium succinate (OVX + MPSS) and 3 sham operated healthy controls) were subjected to radiographic examination before the beginning of the study and after 6 weeks; after sacrifice, they were immediately scanned with a 64-channel CT. A set of morphological features was extracted from RX images and then subjected to Principal Component Analysis (PCA); BMD-values were calculated at different depths from the articular surface of the femur and the tibia. The selected morphological variables and the BMD values were correlated in order to determine useful information in medical diagnostics. RESULTS: Ten morphological variables explained 80.39% of the total variability, but only some of them demonstrated significant differences between baseline and OVX + SSMP states and between OVX + SSMP and control. Spearman analysis showed higher positive and negative correlation of these parameters among them (r = 0.67, 0.81, 0.83 and 1); and a moderate correlation with the BMD values (r = 0.50, 0.52, 0.52 and 0.55) during the progression of osteoporosis (OP). CONCLUSION: The Spearman's rank test supported the overall significant correlation between both, the morphological features and BMD values, making them as a reliable alternative option to the diagnosis of osteoporosis.