Construction of statistical shape atlases for bone structures based on a two-level framework.

BACKGROUND: The statistical shape atlas is a 3D medical image analysis tool that encodes shape variations between populations. However, efficiency, accuracy and finding the correct correspondence are still unsolved issues during the construction of the atlas. METHODS: We developed a two-level-based framework that speeds up the registration process while maintaining accuracy of the atlas. We also proposed a semi-automatic strategy to achieve segmentation and registration simultaneously, without knowing any prior information about the shape. RESULTS: We have separately constructed the atlas for the femur and spine. The experimental results demonstrate the efficiency and accuracy of our methods. CONCLUSIONS: Our two-level framework and semi-automatic strategy are able to efficiently construct the atlas for bone structures without losing accuracy. We can handle either 3D surface data or raw DICOM images.

Fully automated computer algorithm for calculating articular contact points with application to knee biomechanics.

A fully automated computer algorithm for calculating the articular contact points between two bone surface models is presented. The algorithm requires the bone surface models and their relative positions as inputs in order to resolve the articular contact path. In the case of surface model overlap due to measurement errors or as a solution of an optimization procedure, the result is a volumetric estimation of the space confined between the two surfaces. The algorithm is based on attaching a grid of lines to one bone surface model and calculating the intersecting points of each of the lines in the grid with both bone surface models. The contact points are then determined as the closest points between the surfaces along the lines in the grid. The same contact points are used to evaluate any volume that is confined between two overlapping surface models. The algorithm is ideal for use in biomechanical studies, simulations of joint motion, and optimizations that require an iterative process to determine contact path and relative bone position. The algorithm is applied to a Sawbones knee model that is moved from flexion to extension while being tracked by an optical tracking system. The contact path of the two bones is generated and an example of calculating bone impingement is provided.

Computer-assisted evaluation of kinematics of the two bundles of the anterior cruciate ligament.

The aim of this cadaveric study was to describe the kinematics of the anterior cruciate ligament (ACL)-intact, posterolateral (PL) bundle-deficient and ACL-deficient knee by applying a protocol for computer-assisted evaluation of knee kinematics. The hypothesis that the PL bundle functions mainly at low knee flexion angles was tested. An optical tracking system was used to acquire knee joint motion on 10 knees during clinical evaluations by tracking markers rigidly attached to the bones. The protocol included acquisition of anterior-posterior (AP) translations and internal-external (IE) rotations, and evaluation of three clinical knee laxity tests (anterior drawer, manual and instrumented Lachman). The data demonstrated no significant contribution to AP translation and IE laxity from the PL bundle over the entire range of motion. The clinical knee laxity tests showed no significant differences between the ACL-intact and PL bundle-deficient states. The hypothesis could not be proven. Current clinical knee laxity measurements may not be suited for detecting subtle changes such as PL bundle deficiency in the ACL anatomy. The computation of knee laxity might be a step towards a more precise kinematic test of knee stability not only in the native and torn ACL state of the knee but also in the reconstructed knee.

Computer evaluation of kinematics of anterior cruciate ligament reconstructions.

Current clinical and instrumented outcome measurements of knee instability lack accuracy, especially when multiplanar instability is considered. The aim of our cadaveric study was to describe the kinematics in the intact, double bundle, and anteromedial bundle reconstructed anterior cruciate ligament knee by applying a protocol for computer-assisted evaluation of knee kinematics. An optical navigation system was used to acquire knee motion (n = 5) during clinical evaluations by tracking markers rigidly attached to the bones. The protocol included acquisition of anteroposterior translations and internal-external rotations and evaluation of three clinical knee laxity tests (anterior drawer, manual, and instrumented Lachman). Our anteroposterior translation data showed the double-bundle technique and anteromedial bundle technique could restore anteroposterior stability comparable to the intact state. For internal-external laxity, the double-bundle technique demonstrated overcorrection at 15 degrees, 60 degrees, 75 degrees, and 90 degrees. The anterior drawer and manual Lachman knee laxity tests showed improved stability for the double-bundle compared to the anteromedial bundle technique. This pilot study suggests the computation of knee laxity with a high precision method might be a step toward a more precise kinematic test of knee stability for evaluating different reconstruction methods.