A new dynamic RSA set-up for 3D joint motion analysis

Introduction Dynamic roentgen stereophotogrammetric analysis (RSA) showed to be a very accurate method in detecting 3D joint motion1, but at present this technique is based only on custom radiographic equipments. We tested a new dynamic RSA set-up, based on the use of a commercial bi-plane fluoroscopy system. This set-up permits the acquisition of slow passive and active movements of the lower limb, both in supine and weight-bearing position. This method relies on a fully 3D data acquisition, there- fore a more homogeneous accuracy in all the directions of motion is to be expected, in comparison with single plane fluoroscopy 2,3 which is characterized by a large out-of-plane error. Methods We used a Biplane Advantx (GE) system, with two 320mm diameter image intensifiers. Image sequences were recorded at 4fps. We utilized a bi-planar calibration cage and a regular grid for the calibration of systematic geometric distortion. All calibration tools were custom made. Custom toolbox (in MATLAB language) were designed for the correction of distortion following the global bi-polynomial technique4; a commercial software (Model-Based RSA 3.0, Medis Specials bv) were used for the reconstruction of 3D position of the markers; cus- tom software were used for kinematics elaboration. We performed in-vitro tests with a phantom and in-vivo tests examining slow passive motions of 5 patients (3 ligament reconstructions, 2 knee replacements). Tantalum markers were implanted in both femoral and tibial epiphyses of each knee. Bias and precision were investigated in terms of errors in dynamic tracking of markers1 and in terms of translational and rotational accuracy3. Results Dynamic tracking of markers inserted in the phantom object showed a bias of 0.40mm and an inter-marker distance standard deviation of 0.3mm. In-vivo tests showed an average standard deviation of inter-marker distances of 0.3mm, too. From in-vitro tests the mean error in detecting relative motion were not negligible, whereas translational precision was below 0.5mm in all directions and the rotational precision was below 0.3º (Tab.1) Discussion This study shows the potentiality of a dynamic RSA set- up using an available commercial bi-plane fluoroscopy system to provide an interesting method for fully 3D joint motion analysis. Further improvements in terms of reduction of bias of the system may be reached by the use of more precise calibration tools and of different commercial bi-plane fluoroscopic systems, with more suitable specifications for kinematics studies (e.g. with a stricter synchronization of bi-plane acquisitions). References 1. Tashman, S., Anderst, W. In-vivo measurement of dynamic joint motion using high speed biplane radiography and CT: application to canine ACL deficiency. J Biomech Eng, 125, 238, 2003. 2. Hoff, W.A., Komistek, R.D., Dennis, D.A., Gabriel S.M., Walker S.A. Three-dimensional determination of femoral-tibial contact positions under in vivo conditions using fluoroscopy. Clin Biomech, 13, 455, 1998. 3. Garling, E.H., Kaptein, B.L., Geleijns, K., Nelissen, R.G., Valstar, E.R. Marker Configuration Model-Based Roentgen Fluoroscopic Analysis. J Biomech, 38, 893, 2005. 4. Gronenschild E. The accuracy and reproducibility of a global method to correct for geometric image distortion in the x-ray imaging chain. Med Phys, 24, 1875, 1997