A New Angle on Bones: Robust Pose Estimation in X-Ray and Ultrasound
2026-06-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created a method to automatically measure angles between bones in medical images, which helps doctors diagnose and plan treatments faster and more consistently. They use a combination of learning algorithms to find points, then fit lines using techniques that can handle errors well. Their method was tested on three child-related medical tasks and produced results close to what doctors typically observe. It worked better than older methods that relied on specific landmarks. The authors also shared their software and data online for others to use.
bone pose estimationmedical image analysislearning-based point proposalline fittingRANSACHough transformfracture angle measurementdevelopmental dysplasia of the hipultrasound imagingradiographs
Authors
Ron Keuth, Christoph Großbröhmer, Franziska Halm, Miriam Johann, Anne-Nele Schröder, Ludger Tüshaus, Mattias P. Heinrich, Lasse Hansen
Abstract
Measuring the angle between bone structures is a routine task in medical image analysis and provides a key quantitative parameter for diagnosis and treatment planning. Automated methods can reduce time and cost while improving reproducibility. In this work, we address automatic bone pose estimation using a learning-based point candidate proposal followed by a line model to extract axis parameters. Since conventional line models such as least squares are sensitive to outliers, we incorporate false-positive reduction strategies and robust fitting techniques, such as RANSAC and Hough transforms, to improve robustness. We evaluate our method on three clinically relevant paediatric angle estimation tasks: fracture fragment assessment in radiographs and ultrasound and developmental dysplasia of the hip evaluation in ultrasound using the Graf method. Our approach achieves mean errors of $4.1^\circ$, $5.4^\circ$, and $5.51^\circ$, respectively, not only remaining within the expected clinical observer variability, but also significantly outperforming landmark-based methods. Our code and annotations for fracture angle assessment in radiographs are publicly available on GitHub.