This study aimed to develop and validate a fully automated system that quantifies proximal femoral bone mineral density (BMD) from CT images. The study analyzed 978 pairs of hip CT and dual-energy X-ray absorptiometry (DXA) measurements of the proximal femur (DXA-BMD) collected from three institutions. From the CT images, the femur and a calibration phantom were automatically segmented using previously trained deep-learning models. The Hounsfield units of each voxel were converted into density (mg/cm3). Then, a deep-learning model trained by manual landmark selection of 315 cases was developed to select the landmarks at the proximal femur to rotate the CT volume to the neutral position. Finally, the CT volume of the femur was projected onto the coronal plane, and the areal BMD of the proximal femur (CT-aBMD) was quantified. CT-aBMD correlated to DXA-BMD, and a receiver operating characteristic (ROC) analysis quantified the accuracy in diagnosing osteoporosis.Aims
Methods
We have developed a CT-based navigation system using infrared light-emitting diode markers and an optical camera. We used this system to perform cementless total hip replacement using a ceramic-on-ceramic bearing couple in 53 patients (60 hips) between 1998 and 2001. We reviewed 52 patients (59 hips) at a mean of six years (5 to 8) postoperatively. The mid-term results of total hip replacement using navigation were compared with those of 91 patients (111 hips) who underwent this procedure using the same implants, during the same period, without navigation. There were no significant differences in age, gender, diagnosis, height, weight, body mass index, or pre-operative clinical score between the two groups. The operation time was significantly longer where navigation was used, but there was no significant difference in blood loss or navigation-related complications. With navigation, the acetabular components were placed within the safe zone defined by Lewinnek, while without, 31 of the 111 components were placed outside this zone. There was no significant difference in the Merle d’Aubigne and Postel hip score at the final follow-up. However, hips treated without navigation had a higher rate of dislocation. Revision was performed in two cases undertaken without navigation, one for aseptic acetabular loosening and one for fracture of a ceramic liner, both of which showed evidence of neck impingement on the liner. A further five cases undertaken without navigation showed erosion of the posterior aspect of the neck of the femoral component on the lateral radiographs. These seven impingement-related mechanical problems correlated with malorientation of the acetabular component. There were no such mechanical problems in the navigated group. We conclude that CT-based navigation increased the precision of orientation of the acetabular component and control of limb length in total hip replacement, without navigation-related complications. It also reduced the rate of dislocation and mechanical problems related to impingement.
