Abstract:Background: Ante-inclination (AI) of the cup is a key component of the combined sagittal index (CSI) for predicting joint stability after total hip arthroplasty (THA). We aimed to (1) validate a mathematical algorithm relating AI to radiographic anteversion (RA), radiographic inclination (RI), and pelvic tilt (PT); (2) convert the AI criterion of the CSI into the coronal functional safe zone (CFSZ) and explore the influences of standing-to-sitting pelvic motion (PM) and pelvic incidence (PI) on the CFSZ; and (3) attempt to locate a universal cup orientation that always fulfills the AI criterion of the CSI for all patients. Methods: In the first phase, a phantom pelvis was designed to simulate a range of PT values, and an acetabular cup was implanted with different RA, RI, and PT settings using a robot-assisted technique and scanned using the EOS imaging system. The second phase involved patient radiographs. We enrolled 100 patients who underwent robot-assisted THA from April 2019 to December 2019, and EOS images before THA and at the 12-month follow-up were recorded. The AI was measured on a lateral radiograph; this angle was used as the reference and compared with the calculated AI to assess the accuracy of the algorithm. Linear regression was conducted to explore the relationship between the size of the CFSZ and the values of PM and PI. We calculated the intersection of the CFSZs of the patients to find a universal cup orientation (RA and RI) for the CSI. Results: The algorithm was accurate according to both the phantom study and patient radiographs using PT at the time of follow-up, with mean absolute errors (MAEs) of 1.5° (width of 95% confidence interval [CI], 2.2°) and 2.8° (width of 95% CI, 3.0°), respectively. However, the preoperatively predicted AI had an MAE of 9.0° (width of 95% CI, 10.5°). In patient subgroups with lower PM or PI, the sizes of the CFSZ and of its intersection with the Lewinnek safe zone were significantly smaller (p < 0.017). No universal cup orientation existed to fulfill the CSI criteria for all patients or for any of the PM or PI subgroups. Conclusions: The target orientation for the cup in THA should be individualized. Our validated algorithm may serve as a quantitative tool for the patient-specific optimization of cup AI on the basis of the functional safe zone. Clinical relevance: The Lewinnek safe zone fails because it cannot predict the functional orientation of the cup. The concept of a universal safe zone of cup orientation should be abandoned and replaced by a patient-specific surgical target.

Estimation of Pelvic Sagittal Inclination from Anteroposterior Radiograph Using Convolutional Neural Networks: Proof-of-Concept Study

Assessment of Pelvic Tilt in Anteroposterior Radiographs by Area Ratio Based on Deep Learning

Concurrent Learning Approach for Estimation of Pelvic Tilt from Anterior-Posterior Radiograph

Reliability and Validity Analysis of Pelvic Sagittal Inclination Calculated by Inverse Cosine Function Method on Pelvic Anteroposterior Radiographs

Estimation of Acetabular Version from Anteroposterior Pelvic Radiograph Employing Deep Learning

Detecting mechanical loosening of total hip replacement implant from plain radiograph using deep convolutional neural network

Detecting total hip replacement prosthesis design on preoperative radiographs using deep convolutional neural network

Anatomical landmark detection on bi-planar radiographs for predicting spinopelvic parameters

An Artificial Neural Network Model Based on Standing Lateral Radiographs for Predicting Sitting Pelvic Tilt in Healthy Adults.

Estimation of 3D Pose of Femurs in Bi-planar Digital Radiographs

Deep Learning Artificial Intelligence Model for Assessment of Hip Dislocation Risk Following Primary Total Hip Arthroplasty From Postoperative Radiographs

Automated osteoporosis classification and T-score prediction using hip radiographs via deep learning algorithm

Predicting the Risk of Total Hip Replacement by Using A Deep Learning Algorithm on Plain Pelvic Radiographs: Diagnostic Study

A novel approach for screening standard anteroposterior pelvic radiographs in children

Artificial Intelligence-Based Surgery Support Model Using Intraoperative Radiographs for Assessing the Acetabular Component Angle

Automating Linear and Angular Measurements for the Hip and Knee After Computed Tomography: Validation of a Three-Stage Deep Learning and Computer Vision-Based Pipeline for Pathoanatomic Assessment

Personalised pose estimation from single-plane moving fluoroscope images using deep convolutional neural networks.

Clinical Significance of Pose Estimation Methods Compared with Radiographic Parameters in Adolescent Patients with Idiopathic Scoliosis

An automated framework for pediatric hip surveillance and severity assessment using radiographs

Conversion of the Sagittal Functional Safe Zone to the Coronal Plane Using a Mathematical Algorithm: The Reason for Failure of the Lewinnek Safe Zone

Comparison of Transfer Learning Models in Pelvic Tilt and Rotation Measurement in Pediatric Anteroposterior Pelvic Radiographs