IMPROVING QUANTITATIVE TREATMENT RESPONSE MONITORING WITH DEFORMABLE IMAGE REGISTRATION

Quantitative imaging with 18F-FDG PET/CT has the potential to provide an in vivo assessment of response to radiotherapy (RT). However, comparing tissue tracer uptake in longitudinal studies is often confounded by variations in patient setup and potential treatment induced gross anatomic changes. These variations make true response monitoring for the same anatomic volume a challenge, not only for tumors, but also for normal organs-at-risk (OAR). The central hypothesis of this study is that more accurate image registration will lead to improved quantitation of tissue response to RT with 18F-FDG PET/CT. Employing an in-house developed “demons” based deformable image registration algorithm, pre-RT tumor and parotid gland volumes can be more accurately mapped to serial functional images. To test the hypothesis, specific aim 1 was designed to analyze whether deformably mapping tumor volumes rather than aligning to bony structures leads to superior tumor response assessment. We found that deformable mapping of the most metabolically avid regions improved response prediction (P<0.05). The positive predictive power for residual disease was 63% compared to 50% for contrast enhanced post-RT CT. Specific aim 2 was designed to use parotid gland standardized uptake value (SUV) as an objective imaging biomarker for salivary toxicity. We found that relative change in parotid gland SUV correlated strongly with salivary toxicity as defined by the RTOG/EORTC late effects analytic scale (Spearman’s ρ = -0.96, P<0.01). Finally, the goal of specific aim 3 was to create a phenomenological dose-SUV response model for the human parotid glands. Utilizing only baseline metabolic function and the planned dose distribution, predicting parotid SUV change or salivary toxicity, based upon specific aim 2, became possible. We found that the predicted and observed parotid SUV relative changes were significantly correlated (Spearman’s ρ = 0.94, P<0.01). The application of deformable image registration to quantitative treatment response monitoring with 18F-FDG PET/CT could have a profound impact on patient management. Accurate and early identification of residual disease may allow for more timely intervention, while the ability to quantify and predict toxicity of normal OAR might permit individualized refinement of radiation treatment plan designs.

Evaluation of deformable image registration for improved 4D CT-derived ventilation for image guided radiotherapy

Recent treatment planning studies have demonstrated the use of physiologic images in radiation therapy treatment planning to identify regions for functional avoidance. This image-guided radiotherapy (IGRT) strategy may reduce the injury and/or functional loss following thoracic radiotherapy. 4D computed tomography (CT), developed for radiotherapy treatment planning, is a relatively new imaging technique that allows the acquisition of a time-varying sequence of 3D CT images of the patient's lungs through the respiratory cycle. Guerrero et al. developed a method to calculate ventilation imaging from 4D CT, which is potentially better suited and more broadly available for IGRT than the current standard imaging methods. The key to extracting function information from 4D CT is the construction of a volumetric deformation field that accurately tracks the motion of the patient's lungs during the respiratory cycle. The spatial accuracy of the displacement field directly impacts the ventilation images; higher spatial registration accuracy will result in less ventilation image artifacts and physiologic inaccuracies. Presently, a consistent methodology for spatial accuracy evaluation of the DIR transformation is lacking. Evaluation of the 4D CT-derived ventilation images will be performed to assess correlation with global measurements of lung ventilation, as well as regional correlation of the distribution of ventilation with the current clinical standard SPECT. This requires a novel framework for both the detailed assessment of an image registration algorithm's performance characteristics as well as quality assurance for spatial accuracy assessment in routine application. Finally, we hypothesize that hypo-ventilated regions, identified on 4D CT ventilation images, will correlate with hypo-perfused regions in lung cancer patients who have obstructive lesions. A prospective imaging trial of patients with locally advanced non-small-cell lung cancer will allow this hypothesis to be tested. These advances are intended to contribute to the validation and clinical implementation of CT-based ventilation imaging in prospective clinical trials, in which the impact of this imaging method on patient outcomes may be tested.

Rigid and Non-rigid Point-based Medical Image Registration

The primary goal of this dissertation is to develop point-based rigid and non-rigid image registration methods that have better accuracy than existing methods. We first present point-based PoIRe, which provides the framework for point-based global rigid registrations. It allows a choice of different search strategies including (a) branch-and-bound, (b) probabilistic hill-climbing, and (c) a novel hybrid method that takes advantage of the best characteristics of the other two methods. We use a robust similarity measure that is insensitive to noise, which is often introduced during feature extraction. We show the robustness of PoIRe using it to register images obtained with an electronic portal imaging device (EPID), which have large amounts of scatter and low contrast. To evaluate PoIRe we used (a) simulated images and (b) images with fiducial markers; PoIRe was extensively tested with 2D EPID images and images generated by 3D Computer Tomography (CT) and Magnetic Resonance (MR) images. PoIRe was also evaluated using benchmark data sets from the blind retrospective evaluation project (RIRE). We show that PoIRe is better than existing methods such as Iterative Closest Point (ICP) and methods based on mutual information. We also present a novel point-based local non-rigid shape registration algorithm. We extend the robust similarity measure used in PoIRe to non-rigid registrations adapting it to a free form deformation (FFD) model and making it robust to local minima, which is a drawback common to existing non-rigid point-based methods. For non-rigid registrations we show that it performs better than existing methods and that is less sensitive to starting conditions. We test our non-rigid registration method using available benchmark data sets for shape registration. Finally, we also explore the extraction of features invariant to changes in perspective and illumination, and explore how they can help improve the accuracy of multi-modal registration. For multimodal registration of EPID-DRR images we present a method based on a local descriptor defined by a vector of complex responses to a circular Gabor filter.

3D-2D image registration by nonlinear regression

We propose a 3D-2D image registration method that relates image features of 2D projection images to the transformation parameters of the 3D image by nonlinear regression. The method is compared with a conventional registration method based on iterative optimization. For evaluation, simulated X-ray images (DRRs) were generated from coronary artery tree models derived from 3D CTA scans. Registration of nine vessel trees was performed, and the alignment quality was measured by the mean target registration error (mTRE). The regression approach was shown to be slightly less accurate, but much more robust than the method based on an iterative optimization approach.

Image registration techniques with multiresolution analysis in satellite oceanography

En aquest article es fa una descripció dels procediments realitzats per enregistrar dues imatges geomètricament, de forma automàtica, si es pren la primera com a imatge de referència. Es comparen els resultats obtinguts mitjançant tres mètodes. El primer mètode és el d’enregistrament clàssic en domini espacial maximitzant la correlació creuada (MCC)[1]. El segon mètode es basa en aplicar l’enregistrament MCC conjuntament amb un anàlisi multiescala a partir de transformades wavelet [2]. El tercer mètode és una variant de l’anterior que es situa a mig camí dels dos. Per cada un dels mètodes s’obté una estimació dels coeficients de la transformació que relaciona les dues imatges. A continuació es transforma per cada cas la segona imatge i es georeferencia respecte la primera. I per acabar es proposen unes mesures quantitatives que permeten discutir i comparar els resultats obtinguts amb cada mètode.

Compression-based Image Registration

Image registration is an important component of image analysis used to align two or more images. In this paper, we present a new framework for image registration based on compression. The basic idea underlying our approach is the conjecture that two images are correctly registered when we can maximally compress one image given the information in the other. The contribution of this paper is twofold. First, we show that the image registration process can be dealt with from the perspective of a compression problem. Second, we demonstrate that the similarity metric, introduced by Li et al., performs well in image registration. Two different versions of the similarity metric have been used: the Kolmogorov version, computed using standard real-world compressors, and the Shannon version, calculated from an estimation of the entropy rate of the images

Medical image registration based on random line sampling

One of the key aspects in 3D-image registration is the computation of the joint intensity histogram. We propose a new approach to compute this histogram using uniformly distributed random lines to sample stochastically the overlapping volume between two 3D-images. The intensity values are captured from the lines at evenly spaced positions, taking an initial random offset different for each line. This method provides us with an accurate, robust and fast mutual information-based registration. The interpolation effects are drastically reduced, due to the stochastic nature of the line generation, and the alignment process is also accelerated. The results obtained show a better performance of the introduced method than the classic computation of the joint histogram

Fast Geodesic Active Fields for Image Registration Based on Splitting and Augmented Lagrangian Approaches.

In this paper, we present an efficient numerical scheme for the recently introduced geodesic active fields (GAF) framework for geometric image registration. This framework considers the registration task as a weighted minimal surface problem. Hence, the data-term and the regularization-term are combined through multiplication in a single, parametrization invariant and geometric cost functional. The multiplicative coupling provides an intrinsic, spatially varying and data-dependent tuning of the regularization strength, and the parametrization invariance allows working with images of nonflat geometry, generally defined on any smoothly parametrizable manifold. The resulting energy-minimizing flow, however, has poor numerical properties. Here, we provide an efficient numerical scheme that uses a splitting approach; data and regularity terms are optimized over two distinct deformation fields that are constrained to be equal via an augmented Lagrangian approach. Our approach is more flexible than standard Gaussian regularization, since one can interpolate freely between isotropic Gaussian and anisotropic TV-like smoothing. In this paper, we compare the geodesic active fields method with the popular Demons method and three more recent state-of-the-art algorithms: NL-optical flow, MRF image registration, and landmark-enhanced large displacement optical flow. Thus, we can show the advantages of the proposed FastGAF method. It compares favorably against Demons, both in terms of registration speed and quality. Over the range of example applications, it also consistently produces results not far from more dedicated state-of-the-art methods, illustrating the flexibility of the proposed framework.

Influence of magnetic field strength and image registration strategy on voxel-based morphometry in a study of Alzheimer's disease.

Multi-centre data repositories like the Alzheimer's Disease Neuroimaging Initiative (ADNI) offer a unique research platform, but pose questions concerning comparability of results when using a range of imaging protocols and data processing algorithms. The variability is mainly due to the non-quantitative character of the widely used structural T1-weighted magnetic resonance (MR) images. Although the stability of the main effect of Alzheimer's disease (AD) on brain structure across platforms and field strength has been addressed in previous studies using multi-site MR images, there are only sparse empirically-based recommendations for processing and analysis of pooled multi-centre structural MR data acquired at different magnetic field strengths (MFS). Aiming to minimise potential systematic bias when using ADNI data we investigate the specific contributions of spatial registration strategies and the impact of MFS on voxel-based morphometry in AD. We perform a whole-brain analysis within the framework of Statistical Parametric Mapping, testing for main effects of various diffeomorphic spatial registration strategies, of MFS and their interaction with disease status. Beyond the confirmation of medial temporal lobe volume loss in AD, we detect a significant impact of spatial registration strategy on estimation of AD related atrophy. Additionally, we report a significant effect of MFS on the assessment of brain anatomy (i) in the cerebellum, (ii) the precentral gyrus and (iii) the thalamus bilaterally, showing no interaction with the disease status. We provide empirical evidence in support of pooling data in multi-centre VBM studies irrespective of disease status or MFS.

Multimodal image registration of the scoliotic torso for surgical planning

Background This paper presents a method that registers MRIs acquired in prone position, with surface topography (TP) and X-ray reconstructions acquired in standing position, in order to obtain a 3D representation of a human torso incorporating the external surface, bone structures, and soft tissues. Methods TP and X-ray data are registered using landmarks. Bone structures are used to register each MRI slice using an articulated model, and the soft tissue is confined to the volume delimited by the trunk and bone surfaces using a constrained thin-plate spline. Results The method is tested on 3 pre-surgical patients with scoliosis and shows a significant improvement, qualitatively and using the Dice similarity coefficient, in fitting the MRI into the standing patient model when compared to rigid and articulated model registration. The determinant of the Jacobian of the registration deformation shows higher variations in the deformation in areas closer to the surface of the torso. Conclusions The novel, resulting 3D full torso model can provide a more complete representation of patient geometry to be incorporated in surgical simulators under development that aim at predicting the effect of scoliosis surgery on the external appearance of the patient’s torso.

Compression-based Image Registration

Image registration is an important component of image analysis used to align two or more images. In this paper, we present a new framework for image registration based on compression. The basic idea underlying our approach is the conjecture that two images are correctly registered when we can maximally compress one image given the information in the other. The contribution of this paper is twofold. First, we show that the image registration process can be dealt with from the perspective of a compression problem. Second, we demonstrate that the similarity metric, introduced by Li et al., performs well in image registration. Two different versions of the similarity metric have been used: the Kolmogorov version, computed using standard real-world compressors, and the Shannon version, calculated from an estimation of the entropy rate of the images

Medical image registration based on random line sampling

One of the key aspects in 3D-image registration is the computation of the joint intensity histogram. We propose a new approach to compute this histogram using uniformly distributed random lines to sample stochastically the overlapping volume between two 3D-images. The intensity values are captured from the lines at evenly spaced positions, taking an initial random offset different for each line. This method provides us with an accurate, robust and fast mutual information-based registration. The interpolation effects are drastically reduced, due to the stochastic nature of the line generation, and the alignment process is also accelerated. The results obtained show a better performance of the introduced method than the classic computation of the joint histogram

Image registration and image interpolation in volume rendering of cross sections

We outline a method for registration of images of cross sections using the concepts of The Generalized Hough Transform (GHT). The approach may be useful in situations where automation should be a concern. To overcome known problems of noise of traditional GHT we have implemented a slight modified version of the basic algorithm. The modification consists of eliminating points of no interest in the process before the application of the accumulation step of the algorithm. This procedure minimizes the amount of accumulation points while reducing the probability of appearing of spurious peaks. Also, we apply image warping techniques to interpolate images among cross sections. This is needed where the distance of samples between sections is too large. Then it is suggested that the step of registration with GHT can help the interpolation automation by simplifying the correspondence between points of images. Some results are shown.

Algorithms for on-line image registration from multiple views

Images of a scene, static or dynamic, are generally acquired at different epochs from different viewpoints. They potentially gather information about the whole scene and its relative motion with respect to the acquisition device. Data from different (in the spatial or temporal domain) visual sources can be fused together to provide a unique consistent representation of the whole scene, even recovering the third dimension, permitting a more complete understanding of the scene content. Moreover, the pose of the acquisition device can be achieved by estimating the relative motion parameters linking different views, thus providing localization information for automatic guidance purposes. Image registration is based on the use of pattern recognition techniques to match among corresponding parts of different views of the acquired scene. Depending on hypotheses or prior information about the sensor model, the motion model and/or the scene model, this information can be used to estimate global or local geometrical mapping functions between different images or different parts of them. These mapping functions contain relative motion parameters between the scene and the sensor(s) and can be used to integrate accordingly informations coming from the different sources to build a wider or even augmented representation of the scene. Accordingly, for their scene reconstruction and pose estimation capabilities, nowadays image registration techniques from multiple views are increasingly stirring up the interest of the scientific and industrial community. Depending on the applicative domain, accuracy, robustness, and computational payload of the algorithms represent important issues to be addressed and generally a trade-off among them has to be reached. Moreover, on-line performance is desirable in order to guarantee the direct interaction of the vision device with human actors or control systems. This thesis follows a general research approach to cope with these issues, almost independently from the scene content, under the constraint of rigid motions. This approach has been motivated by the portability to very different domains as a very desirable property to achieve. A general image registration approach suitable for on-line applications has been devised and assessed through two challenging case studies in different applicative domains. The first case study regards scene reconstruction through on-line mosaicing of optical microscopy cell images acquired with non automated equipment, while moving manually the microscope holder. By registering the images the field of view of the microscope can be widened, preserving the resolution while reconstructing the whole cell culture and permitting the microscopist to interactively explore the cell culture. In the second case study, the registration of terrestrial satellite images acquired by a camera integral with the satellite is utilized to estimate its three-dimensional orientation from visual data, for automatic guidance purposes. Critical aspects of these applications are emphasized and the choices adopted are motivated accordingly. Results are discussed in view of promising future developments.