An interpolation technique to enable accurate three-dimensional joint kinematic analyses using asynchronous biplane fluoroscopy. (October 2018)

• Record Type: Journal Article
• Authors: Akbari-Shandiz, Mohsen; Mozingo, Joseph D.; Holmes III, David R.; Zhao, Kristin D.
• Is Part Of: Medical engineering & physics. Volume 60(2018)
• Page Start: 109
• Page End: 116
• Publication Date: 2018-10
• Subjects: Clinical biplane fluoroscopic systems -- Asynchronous image acquisition -- Kinematics -- 2D-3D model-based registration -- Bead tracking -- Interpolation; Asynch asynchronous -- DRR digitally reconstructed radiograph -- FP frontal plane -- SP sagittal plane -- GH glenohumeral; Biomedical engineering -- Periodicals; Biomedical Engineering -- Periodicals; Physics -- Periodicals; Génie biomédical -- Périodiques; Biomedical engineering; Electronic journals; Periodicals; 610.28
• DOI: 10.1016/j.medengphy.2018.07.007
• Languages: English
• ISSNs: 1350-4533

Abstract

Highlights: Asynchronous clinical biplane systems introduce errors to model-based registration. An interpolation technique is introduced to generate synchronous image estimates. A phantom and shoulder studies are used to evaluate improvement in kinematic accuracy. The interpolated registration was in better agreement with the gold standard. It is particularly useful for improving kinematic accuracy of high velocity motions. Abstract: Biplane 2D-3D model-based registration and radiostereometric analysis (RSA) approaches have been commonly used for measuring three-dimensional, in vivo joint kinematics. However, in clinical biplane systems, the x-ray images are acquired asynchronously, which introduces registration errors. The present study introduces an interpolation technique to reduce image registration error by generating synchronous fluoroscopy image estimates. A phantom study and cadaveric shoulder study were used to evaluate the level of improvement in image registration that could be obtained as a result of using our interpolation technique. Our phantom study results show that the interpolated bead tracking technique was in better agreement with the true bead positions than when asynchronous images were used alone. The overall RMS error of glenohumeral kinematics for interpolated biplane registration was reduced by 1.27 mm, 0.40 mm, and 0.47 mm in anterior-posterior, superior-inferior, and medial-lateral translation, respectively; and 0.47°, 0.67°, and 0.19° in ab-adduction, internal-external rotation and flexion-extension, respectively, compared to asynchronous registration. The interpolated biplane registration results were consistent with previously reported studies using custom synchronous biplane fluoroscopy technology. This approach will be particularly useful for improving the kinematic accuracy of high velocity activities when using clinical biplane fluoroscopes or two independent c-arms, which are available at a number of institutions.