Hindfoot disorders are complex 3D deformities. Current literature has assessed their influence on the full leg alignment, but the superposition of the hindfoot on plain radiographs resulted in different measurement errors. Therefore, the aim of this study is to assess the hindfoot alignment on Weight-Bearing CT (WBCT) and its influence on the radiographic Hip-Knee-Ankle (HKA) angle. A retrospective analysis was performed on a study population of 109 patients (mean age of 53 years ± 14,49) with a varus or valgus hindfoot deformity. The hindfoot angle (HA) was measured on the WBCT while the HKA angle, and the anatomical tibia axis angle towards the vertical (TA. X. ) were analysed on the Full Leg radiographs. The mean HA in the valgus hindfoot group was 9,19°±7.94, in the varus hindfoot group −7,29°±6.09. The mean TA. X. was 3,32°±2.17 in the group with a valgus hindfoot and 1,89°±2.63 in the group with a varus hindfoot, which showed to be statistically different (p<0.05). The mean HKA Angle was −1,35°±2.73 in the valgus hindfoot group and 0,4°±2.89 in the varus hindfoot group, which showed to be statistically different (p<0.05). This study demonstrates a higher varus in both the HKA and TA. X. in valgus hindfoot and a higher tibia valgus in varus hindfoot. This contradicts the previous assumption that a varus hindfoot is associated with a varus knee or vice versa. In clinical practice, these findings contribute to a better understanding of deformity corrections of both the hindfoot and the knee

Abstract. Objectives. Knee alignment affects both the development and surgical treatment of knee osteoarthritis. Automating femorotibial angle (FTA) and hip-knee-ankle angle (HKA) measurement from radiographs could improve reliability and save time. Further, if the gold-standard HKA from full-limb radiographs could be accurately predicted from knee-only radiographs then the need for more expensive equipment and radiation exposure could be reduced. The aim of this research is to assess if deep learning methods can predict FTA and HKA angle from posteroanterior (PA) knee radiographs. Methods. Convolutional neural networks with densely connected final layers were trained to analyse PA knee radiographs from the Osteoarthritis Initiative (OAI) database with corresponding angle measurements. The FTA dataset with 6149 radiographs and HKA dataset with 2351 radiographs were split into training, validation and test datasets in a 70:15:15 ratio. Separate models were learnt for the prediction of FTA and HKA, which were trained using mean squared error as a loss function. Heat maps were used to identify the anatomical features within each image that most contributed to the predicted angles. Results. FTA could be predicted with errors less than 3° for 99.8% of images, and less than 1° for 89.5%. HKA prediction was less accurate than FTA but still high: 95.7% within 3°, and 68.0 % within 1°. Heat maps for both models were generally concentrated on the knee anatomy and could prove a valuable tool for assessing prediction reliability in clinical application. Conclusions. Deep learning techniques could enable fast, reliable and accurate predictions of both FTA and HKA from plain knee radiographs. This could lead to cost savings for healthcare providers and reduced radiation exposure for patients

Introduction. With the development of 3D printing technology, there are many different types of PSI in the world. The accuracy of patient specific instrumentation (PSI) in primary total knee arthroplasty (TKA) is dependent on appropriate placement of the cutting blocks. However, previous reports on one type of PSI measured the difference between postoperative prosthetic alignment and postoperative mechanical axis and thus these reports did not evaluate intraoperative comparison of PSIs between two different designs. The purpose of this study was to evaluate the intraoperative accuracy of two different designed PSIs (My knee, Medacta International, Castel San Pietro, Switzerland) with two examiners using CT free navigation system (Stryker, Mahwar, NJ, USA) in regards to sagittal and coronal alignment. Methods. We enrolled 78knees (66 patients) with a primary cemented TKA using two different designed CT-based PSIs (My knee, Medacta International, Castel San Pietro, Switzerland). All operations were performed by two senior surgeons who have experience with greater than 500 TKAs and greater than 200 navigated TKAs. Two examiners were same two surgeons. The study period was between June 2015 and November 2016. The local ethics' committee approved the study prior to its initiation, and informed consent was obtained from all patients. After placement of the PSI on the femur and tibia, the position of the PSI was evaluated by s intraoperative navigation. Two examiners placed two different types (STD(standard) and MIS(minimum invasive surgery)) of PSI on same joint. As required by the PSI, only soft- tissue was removed and osteophytes were left in place. Femoral MIS PSI was required partial remove of lateral cartilage. For the femur, the coronal position in relation to the mechanical axis were documented. For the tibia, the coronal alignment and the tibial slope were documented. Of note, intraoperative modifications to the PSI were not made based upon the results of the navigation. Rather, the findings of the intraoperative navigation were simply documented. Results. The mean age of the cohort was 72.9±7.5years (range, 55–85years). The study included 11men and 55women, with a mean height of 151±8.2cm (range, 135–175cm), mean weight of 59.4±4.3kg (range, 42–82kg), and a mean of Body Mass Index of 25.9±3.6 (range, 17.2–36.4). HKA angle (supine position) measured by CT was 170.8 ±4.4 degree(range, 162.5–182degree). Diagnosis was osteoarthritis in all patient. There was no statistically significant difference in PSI position alignment for femoral flexion, tibial coronal angle, tibial slope between the two groups with two examiners. However, the intraoperative coronal position using the femoral STD PSI significantly deviated from using femoral MIS PSI from both examiners. (PSI vs. MIS, examiner1 p = 0.02, examiner2 p=0.04)