High-resolution temporal reconstruction of ankle joint from dynamic MRI

Abstract

Cerebral palsy is the leading cause of motor disabilities affecting children. The ankle is the most common equine musculoskeletal strain in children with cerebral palsy. Despite multiple medical and surgical therapies, postoperative recurrence rate is still very high (48%). A major reason for therapy failure is the lack of knowledge of the ankle joint biomechanics. Dynamic MRI can be used to acquire high resolution static data and low resolution temporal images. However, spatial and temporal data should be combined to provide the most comprehensive point of view to study joint motion. In this paper, we first present an intensity-based registration method to accurately estimate the rigid motion of the ankle bones. Second, we investigate the use of the log-euclidean framework to reconstruct a four-dimensional (3D+time) high-resolution dynamic MRI sequence from a low-resolution dynamic sequence and one high resolution static MR image. The proposed approach has been applied and evaluated on in vivo MRI data acquired for a pilot study on child motor disability. Results demonstrate the robustness of the proposed pipeline and very promising high resolution visualization of the ankle joint.

A fast and memory-efficient algorithm for smooth interpolation of polyrigid transformations: application to human joint tracking

“…However, the use of this method results in excessive memory requirements as it requires multiple matrix multiplications (see Figure 2) and thus it is not suitable for estimating or interpolating very dense deformation fields (e.g. when registering very high-resolution MRI data [19]). To overcome this limitation, we have proposed to compute the matrix exponential using matrix eigendecomposition.…”

4D in vivo quantification of ankle joint space width using dynamic MRI

“…The accuracies of bone motion estimation have been presented in our previous work using DICE score [9]. In this work, we have also quantified the accuracy of joint space area tracking.…”

“…This algorithm require only the manual segmentations of bones of interest in the high-resolution static image as inputs. For more details about this algorithm, the reader is referred to our previous works [9], [11].…”

“…In this paper, we extend these techniques to the measurement of the JSW during the entire joint motion cycle using dynamic MRI sequences, which will certainly provide new temporal motion features. In our previous work [9], we have proposed a motion-based algorithm for estimating HR dynamic MRI sequences via a log-euclidean polyrigid framework (LEPF) [10] using both static and dynamic MRI. In the current work, we use the associated motion estimation algorithm to track the tibiotalar joint with high accuracy and without the need for manual segmentation of dynamic data.…”

A new geodesic-based feature for characterization of 3D shapes: application to soft tissue organ temporal deformations

“…To overcome these limitations, a combination of spatial resolution of conventional static MRI and temporal resolution of dynamic MRI data is necessary. First studies for estimating dense deformation fields from a static MRI scan to a set of low-resolution dynamic MRI scans are presented in [17,16,18] in the context of articulated polyrigid registration of human joints. Another study to perform a high-resolution temporal reconstruction of the bladder during loading exercises is recently introduced in [21].…”