Related Experiment Video
Updated: Aug 7, 2025

In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
Published on: July 2, 2021
Artificial Intelligence Autonomously Measures Cup Orientation, Corrects for Pelvis Orientation, and Identifies
Michael P Murphy1, Cameron J Killen1, Sara R Winfrey1
1Department of Orthopaedic Surgery and Rehabilitation, Loyola University Medical Center, Maywood, Illinois.
An artificial intelligence (AI) program accurately measures hip implant cup orientation from X-rays, outperforming manual methods and identifying retroverted cups. This AI tool offers a faster, more precise approach for total hip arthroplasty (THA) assessments.
Area of Science:
- Orthopedic surgery
- Medical imaging
- Artificial intelligence in medicine
Background:
- Cup orientation in total hip arthroplasty (THA) is critical for preventing impingement and dislocation.
- Manual measurement of cup orientation from radiographs is time-consuming and inaccurate.
- Accurate assessment of cup orientation is essential for successful THA outcomes.
Purpose of the Study:
- To develop and validate an artificial intelligence (AI) program for autonomous measurement of cup orientation from antero-posterior pelvic radiographs.
- To assess the AI program's ability to correct for pelvic orientation and identify cup retroversion.
- To compare the accuracy and efficiency of AI-based measurements against traditional manual methods.
Main Methods:
- Utilized 504 computed tomographic (CT) scans from 2,945 patients undergoing THA for 3D reconstruction and cup orientation measurement.
- Developed an AI model trained on 4,000 X-rays, with data augmentation creating 4,000,000 training images.
- Validated the AI model's accuracy against CT measurements using a test group of 690 radiographs.
Main Results:
- AI predictions for cup orientation were obtained rapidly, averaging 0.22 ± 0.03 seconds per radiograph.
- AI measurements demonstrated high correlation with CT scans (Pearson's r = 0.976 for anteversion, 0.984 for inclination), significantly outperforming manual measurements (r = 0.650 and 0.687).
- The AI model achieved 100.0% accuracy in identifying retroverted cups (17 cases) from single antero-posterior radiographs.
Conclusions:
- AI algorithms can accurately measure cup orientation from radiographs, correcting for pelvic tilt and surpassing manual measurement accuracy.
- This AI approach offers a timely and efficient solution for assessing cup orientation in THA.
- This represents the first AI-driven method capable of identifying retroverted cups using a single antero-posterior radiograph.
More Related Videos
05:57A Teleoperated Robotic System-Assisted Percutaneous Transiliac-Transsacral Screw Fixation Technique
Published on: January 6, 2023
06:28Biomechanical Changes Related to Low Back Pain: An Innovative Tool for Movement Pattern Assessment and Treatment Evaluation in Rehabilitation
Published on: December 13, 2024
Related Concept Videos
Anatomical Positions
The body is upright, facing forward, and standing erect.
The feet are parallel and flat on the floor.
The arms are hanging by the...
Imaging Studies V: Intravenous Urography and Retrograde Pyelography