Training models of anatomic shape variability

A robust multi-variability model based liver segmentation algorithm for CT-scan and MRI modalities

Developing methods to segment the liver in medical images, study and analyze it remains a significant challenge. The shape of the liver can vary considerably from one patient to another, and adjacent organs are visualized in medical images with similar intensities, making the boundaries of the liver ambiguous. Consequently, automatic or semi-automatic segmentation of liver is a difficult task. Moreover, scanning systems and magnetic resonance imaging have different settings and parameters. Thus the images obtained differ from one machine to another. In this article, we propose an automatic model-based segmentation that allows building a faithful 3-D representation of the liver, with a mean Dice value equal to 90.3% on CT and MRI datasets. We compare our algorithm with a semi-automatic method and with other approaches according to the state of the art. Our method works with different data sources, we use a large quantity of CT and MRI images from machines in various hospitals and multiple DICOM images available from public challenges. Finally, for evaluation of liver segmentation approaches in state of the art, robustness is not adequacy addressed with a precise definition. Another originality of this article is the introduction of a novel measure of robustness, which takes into account the liver variability at different scales.

Use of regularized principal component analysis to model anatomical changes during head and neck radiation therapy for treatment adaptation and response assessment

PURPOSE

To develop standard (SPCA) and regularized (RPCA) principal component analysis models of anatomical changes from daily cone beam CTs (CBCTs) of head and neck (H&N) patients and assess their potential use in adaptive radiation therapy, and for extracting quantitative information for treatment response assessment.

METHODS

Planning CT images of ten H&N patients were artificially deformed to create "digital phantom" images, which modeled systematic anatomical changes during radiation therapy. Artificial deformations closely mirrored patients' actual deformations and were interpolated to generate 35 synthetic CBCTs, representing evolving anatomy over 35 fractions. Deformation vector fields (DVFs) were acquired between pCT and synthetic CBCTs (i.e., digital phantoms) and between pCT and clinical CBCTs. Patient-specific SPCA and RPCA models were built from these synthetic and clinical DVF sets. EigenDVFs (EDVFs) having the largest eigenvalues were hypothesized to capture the major anatomical deformations during treatment.

RESULTS

Principal component analysis (PCA) models achieve variable results, depending on the size and location of anatomical change. Random changes prevent or degrade PCA's ability to detect underlying systematic change. RPCA is able to detect smaller systematic changes against the background of random fraction-to-fraction changes and is therefore more successful than SPCA at capturing systematic changes early in treatment. SPCA models were less successful at modeling systematic changes in clinical patient images, which contain a wider range of random motion than synthetic CBCTs, while the regularized approach was able to extract major modes of motion.

CONCLUSIONS

Leading EDVFs from the both PCA approaches have the potential to capture systematic anatomical change during H&N radiotherapy when systematic changes are large enough with respect to random fraction-to-fraction changes. In all cases the RPCA approach appears to be more reliable at capturing systematic changes, enabling dosimetric consequences to be projected once trends are established early in a treatment course, or based on population models.

Non-Euclidean classification of medically imaged objects via s-reps

Classifying medically imaged objects, e.g., into diseased and normal classes, has been one of the important goals in medical imaging. We propose a novel classification scheme that uses a skeletal representation to provide rich non-Euclidean geometric object properties. Our statistical method combines distance weighted discrimination (DWD) with a carefully chosen Euclideanization which takes full advantage of the geometry of the manifold on which these non-Euclidean geometric object properties (GOPs) live. Our method is evaluated via the task of classifying 3D hippocampi between schizophrenics and healthy controls. We address three central questions. 1) Does adding shape features increase discriminative power over the more standard classification based only on global volume? 2) If so, does our skeletal representation provide greater discriminative power than a conventional boundary point distribution model (PDM)? 3) Especially, is Euclideanization of non-Euclidean shape properties important in achieving high discriminative power? Measuring the capability of a method in terms of area under the receiver operator characteristic (ROC) curve, we show that our proposed method achieves strongly better classification than both the classification method based on global volume alone and the s-rep-based classification method without proper Euclideanization of non-Euclidean GOPs. We show classification using Euclideanized s-reps is also superior to classification using PDMs, whether the PDMs are first Euclideanized or not. We also show improved performance with Euclideanized boundary PDMs over non-linear boundary PDMs. This demonstrates the benefit that proper Euclideanization of non-Euclidean GOPs brings not only to s-rep-based classification but also to PDM-based classification.

Incorporating population-level variability in orthopedic biomechanical analysis: a review

Effectively addressing population-level variability within orthopedic analyses requires robust data sets that span the target population and can be greatly facilitated by statistical methods for incorporating such data into functional biomechanical models. Data sets continue to be disseminated that include not just anatomical information but also key mechanical data including tissue or joint stiffness, gait patterns, and other inputs relevant to analysis of joint function across a range of anatomies and physiologies. Statistical modeling can be used to establish correlations between a variety of structural and functional biometrics rooted in these data and to quantify how these correlations change from health to disease and, finally, to joint reconstruction or other clinical intervention. Principal component analysis provides a basis for effectively and efficiently integrating variability in anatomy, tissue properties, joint kinetics, and kinematics into mechanistic models of joint function. With such models, bioengineers are able to study the effects of variability on biomechanical performance, not just on a patient-specific basis but in a way that may be predictive of a larger patient population. The goal of this paper is to demonstrate the broad use of statistical modeling within orthopedics and to discuss ways to continue to leverage these techniques to improve biomechanical understanding of orthopedic systems across populations.

A semiautomatic tool for prostate segmentation in radiotherapy treatment planning

BACKGROUND

Delineation of the target volume is a time-consuming task in radiotherapy treatment planning, yet essential for a successful treatment of cancers such as prostate cancer. To facilitate the delineation procedure, the paper proposes an intuitive approach for 3D modeling of the prostate by slice-wise best fitting ellipses.

METHODS

The proposed estimate is initialized by the definition of a few control points in a new patient. The method is not restricted to particular image modalities but assumes a smooth shape with elliptic cross sections of the object. A training data set of 23 patients was used to calculate a prior shape model. The mean shape model was evaluated based on the manual contour of 10 test patients. The patient records of training and test data are based on axial T1-weighted 3D fast-field echo (FFE) sequences. The manual contours were considered as the reference model. Volume overlap (Vo), accuracy (Ac) (both ratio, range 0-1, optimal value 1) and Hausdorff distance (HD) (mm, optimal value 0) were calculated as evaluation parameters.

RESULTS

The median and median absolute deviation (MAD) between manual delineation and deformed mean best fitting ellipses (MBFE) was Vo (0.9 ± 0.02), Ac (0.81 ± 0.03) and HD (4.05 ± 1.3)mm and between manual delineation and best fitting ellipses (BFE) was Vo (0.96 ± 0.01), Ac (0.92 ± 0.01) and HD (1.6 ± 0.27)mm. Additional results show a moderate improvement of the MBFE results after Monte Carlo Markov Chain (MCMC) method.

CONCLUSIONS

The results emphasize the potential of the proposed method of modeling the prostate by best fitting ellipses. It shows the robustness and reproducibility of the model. A small sample test on 8 patients suggest possible time saving using the model.

Comparison of User-Directed and Automatic Mapping of the Planned Isocenter to Treatment Space for Prostate IGRT

Image-guided radiotherapy (IGRT), adaptive radiotherapy (ART), and online reoptimization rely on accurate mapping of the radiation beam isocenter(s) from planning to treatment space. This mapping involves rigid and/or nonrigid registration of planning (pCT) and intratreatment (tCT) CT images. The purpose of this study was to retrospectively compare a fully automatic approach, including a non-rigid step, against a user-directed rigid method implemented in a clinical IGRT protocol for prostate cancer. Isocenters resulting from automatic and clinical mappings were compared to reference isocenters carefully determined in each tCT. Comparison was based on displacements from the reference isocenters and prostate dose-volume histograms (DVHs). Ten patients with a total of 243 tCTs were investigated. Fully automatic registration was found to be as accurate as the clinical protocol but more precise for all patients. The average of the unsigned x, y, and z offsets and the standard deviations (σ) of the signed offsets computed over all images were (avg. ±  σ (mm)): 1.1 ± 1.4, 1.8 ± 2.3, 2.5 ± 3.5 for the clinical protocol and 0.6 ± 0.8, 1.1 ± 1.5 and 1.1 ± 1.4 for the automatic method. No failures or outliers from automatic mapping were observed, while 8 outliers occurred for the clinical protocol.

Femoral head shape differences during development may identify hips at risk of degeneration

Developmental dysplasia of the hip (DDH) is a common cause of elevated contact stress and early onset osteoarthritis (OA). We hypothesized that adaptation to focal loading during postnatal development would result in signature changes to the shape of the femoral head secondary center of ossification (SCO). SCO shape was evaluated in a canine model of DDH at ages 14 and 32 weeks. The evolving 3D morphology of the SCO was captured using serial quantitative computed tomography. A discrete medial representation shape model was fit to each SCO and served as the basis for quantitative thickness and bending measurements. Shape measurements were tested for associations with hip subluxation and degeneration. At 32 weeks, the SCO was thinner (flatter) in the perifoveal region, the site of focal loading; a greater bend to the SCO was present lateral to the site of thinning; SCO thinning and bending were associated with less femoral head coverage and with a higher probability of degeneration. Shape changes were not detected at 14 weeks. Measurement and visualization of SCO shape changes due to altered loading may provide a basis for identifying hips at risk of early onset OA and a tool for surgical planning of hip restructuring.

A population-based model to describe geometrical uncertainties in radiotherapy: applied to prostate cases

Local motions and deformations of organs between treatment fractions introduce geometrical uncertainties into radiotherapy. These uncertainties are generally taken into account in the treatment planning by enlarging the radiation target by a margin around the clinical target volume. However, a practical method to fully include these uncertainties is still lacking. This paper proposes a model based on the principal component analysis to describe the patient-specific local probability distributions of voxel motions so that the average values and variances of the dose distribution can be calculated and fully used later in inverse treatment planning. As usually only a very limited number of data for new patients is available; in this paper the analysis is extended to use population data. A basic assumption (which is justified retrospectively in this paper) is that general movements and deformations of a specific organ are similar despite variations in the shapes of the organ over the population. A proof of principle of the method for deformations of the prostate and the seminal vesicles is presented.

[Automatic image segmentation for treatment planning in radiotherapy]

One drawback of the growth in conformal radiotherapy and image-guided radiotherapy is the increased time needed to define the volumes of interest. This also results in inter- and intra-observer variability. However, developments in computing and image processing have enabled these tasks to be partially or totally automated. This article will provide a detailed description of the main principles of image segmentation in radiotherapy, its applications and the most recent results in a clinical context.

Image estimation from marker locations for dose calculation in prostate radiation therapy

Tracking implanted markers in the prostate during each radiation treatment delivery provides an accurate approximation of prostate location, which enables the use of higher daily doses with tighter margins of the treatment beams and thus improves the efficiency of the radiotherapy. However, the lack of 3D image data with such a technique prevents calculation of delivered dose as required for adaptive planning. We propose to use a reference statistical shape model generated from the planning image and a deformed version of the reference model fitted to the implanted marker locations during treatment to estimate a regionally dense deformation from the planning space to the treatment space. Our method provides a means of estimating the treatment image by mapping planning image data to treatment space via the deformation field and therefore enables the calculation of dose distributions with marker tracking techniques during each treatment delivery.