Supervised learning for bone shape and cortical thickness estimation from CT images for finite element analysis.

Knowledge about the thickness of the cortical bone is of high interest for fracture risk assessment. Most finite element model solutions overlook this information because of the coarse resolution of the CT images. To circumvent this limitation, a three-steps approach is proposed. 1) Two initial surface meshes approximating the outer and inner cortical surfaces are generated via a shape regression based on morphometric features and statistical shape model parameters. 2) The meshes are then corrected locally using a supervised learning model build from image features extracted from pairs of QCT (0.3-1 mm resolution) and HRpQCT images (82 µm resolution). As the resulting meshes better follow the cortical surfaces, the cortical thickness can be estimated at sub-voxel precision. 3) The meshes are finally regularized by a Gaussian process model featuring a two-kernel model, which seamlessly enables smoothness and shape-awareness priors during regularization. The resulting meshes yield high-quality mesh element properties, suitable for construction of tetrahedral meshes and finite element simulations. This pipeline was applied to 36 pairs of proximal femurs (17 males, 19 females, 76 ±â€¯12 years) scanned under QCT and HRpQCT modalities. On a set of leave-one-out experiments, we quantified accuracy (root mean square error = 0.36 ±â€¯0.29 mm) and robustness (Hausdorff distance = 3.90 ±â€¯1.57 mm) of the outer surface meshes. The error in the estimated cortical thickness (0.05 ±â€¯0.40 mm), and the tetrahedral mesh quality (aspect ratio = 1.4 ±â€¯0.02) are also reported. The proposed pipeline produces finite element meshes with patient-specific bone shape and sub-voxel cortical thickness directly from CT scans. It also ensures that the nodes and elements numbering remains consistent and independent of the morphology, which is a distinct advantage in population studies.

Highly Accurate Facial Nerve Segmentation Refinement From CBCT/CT Imaging Using a Super-Resolution Classification Approach.

Facial nerve segmentation is of considerable importance for preoperative planning of cochlear implantation. However, it is strongly influenced by the relatively low resolution of the cone-beam computed tomography (CBCT) images used in clinical practice. In this paper, we propose a super-resolution classification method, which refines a given initial segmentation of the facial nerve to a subvoxel classification level from CBCT/CT images. The super-resolution classification method learns the mapping from low-resolution CBCT/CT images to high-resolution facial nerve label images, obtained from manual segmentation on micro-CT images. We present preliminary results on dataset, 15 ex vivo samples scanned including pairs of CBCT/CT scans and high-resolution micro-CT scans, with a leave-one-out evaluation, and manual segmentations on micro-CT images as ground truth. Our experiments achieved a segmentation accuracy with a Dice coefficient of , surface-to-surface distance of , and Hausdorff distance of . We compared the proposed technique to two other semi-automated segmentation software tools, ITK-SNAP and GeoS, and show the ability of the proposed approach to yield subvoxel levels of accuracy in delineating the facial nerve.

A novel registration-based methodology for prediction of trabecular bone fabric from clinical QCT: A comprehensive analysis.

Osteoporosis leads to hip fractures in aging populations and is diagnosed by modern medical imaging techniques such as quantitative computed tomography (QCT). Hip fracture sites involve trabecular bone, whose strength is determined by volume fraction and orientation, known as fabric. However, bone fabric cannot be reliably assessed in clinical QCT images of proximal femur. Accordingly, we propose a novel registration-based estimation of bone fabric designed to preserve tensor properties of bone fabric and to map bone fabric by a global and local decomposition of the gradient of a non-rigid image registration transformation. Furthermore, no comprehensive analysis on the critical components of this methodology has been previously conducted. Hence, the aim of this work was to identify the best registration-based strategy to assign bone fabric to the QCT image of a patient's proximal femur. The normalized correlation coefficient and curvature-based regularization were used for image-based registration and the Frobenius norm of the stretch tensor of the local gradient was selected to quantify the distance among the proximal femora in the population. Based on this distance, closest, farthest and mean femora with a distinction of sex were chosen as alternative atlases to evaluate their influence on bone fabric prediction. Second, we analyzed different tensor mapping schemes for bone fabric prediction: identity, rotation-only, rotation and stretch tensor. Third, we investigated the use of a population average fabric atlas. A leave one out (LOO) evaluation study was performed with a dual QCT and HR-pQCT database of 36 pairs of human femora. The quality of the fabric prediction was assessed with three metrics, the tensor norm (TN) error, the degree of anisotropy (DA) error and the angular deviation of the principal tensor direction (PTD). The closest femur atlas (CTP) with a full rotation (CR) for fabric mapping delivered the best results with a TN error of 7.3 ± 0.9%, a DA error of 6.6 ± 1.3% and a PTD error of 25 ± 2°. The closest to the population mean femur atlas (MTP) using the same mapping scheme yielded only slightly higher errors than CTP for substantially less computing efforts. The population average fabric atlas yielded substantially higher errors than the MTP with the CR mapping scheme. Accounting for sex did not bring any significant improvements. The identified fabric mapping methodology will be exploited in patient-specific QCT-based finite element analysis of the proximal femur to improve the prediction of hip fracture risk.

Statistical analysis of the inter-individual variations of the bone shape, volume fraction and fabric and their correlations in the proximal femur.

Including structural information of trabecular bone improves the prediction of bone strength and fracture risk. However, this information is available in clinical CT scans, only for peripheral bones. We hypothesized that a correlation exists between the shape of the bone, its volume fraction (BV/TV) and fabric, which could be characterized using statistical modeling. High-resolution peripheral computed tomography (HR-pQCT) images of 73 proximal femurs were used to build a combined statistical model of shape, BV/TV and fabric. The model was based on correspondence established by image registration and by morphing of a finite element mesh describing the spatial distribution of the bone properties. Results showed no correlation between the distribution of bone shape, BV/TV and fabric. Only the first mode of variation associated with density and orientation showed a strong relationship (R2>0.8). In addition, the model showed that the anisotropic information of the proximal femur does not vary significantly in a population of healthy, osteoporotic and osteopenic samples. In our dataset, the average anisotropy of the population was able to provide a close approximation of the patient-specific anisotropy. These results were confirmed by homogenized finite element (hFE) analyses, which showed that the biomechanical behavior of the proximal femur was not significantly different when the average anisotropic information of the population was used instead of patient-specific fabric extracted from HR-pQCT. Based on these findings, it can be assumed that the fabric information of the proximal femur follows a similar structure in an elderly population of healthy, osteopenic and osteoporotic proximal femurs.

Facial nerve image enhancement from CBCT using supervised learning technique.

Facial nerve segmentation plays an important role in surgical planning of cochlear implantation. Clinically available CBCT images are used for surgical planning. However, its relatively low resolution renders the identification of the facial nerve difficult. In this work, we present a supervised learning approach to enhance facial nerve image information from CBCT. A supervised learning approach based on multi-output random forest was employed to learn the mapping between CBCT and micro-CT images. Evaluation was performed qualitatively and quantitatively by using the predicted image as input for a previously published dedicated facial nerve segmentation, and cochlear implantation surgical planning software, OtoPlan. Results show the potential of the proposed approach to improve facial nerve image quality as imaged by CBCT and to leverage its segmentation using OtoPlan.