Ashraf Mohamed

A framework for predictive modeling of anatomical deformations

A framework for modeling and predicting anatomical deformations is presented, and tested on simulated images. Although a variety of deformations can be modeled in this framework, emphasis is placed on surgical planning, and particularly on modeling and predicting changes of anatomy between preoperative and intraoperative positions, as well as on deformations induced by tumor growth. Two methods are examined. The first is purely shape-based and utilizes the principal modes of co-variation between anatomy and deformation in order to statistically represent deformability. When a patients anatomy is available, it is used in conjunction with the statistical model to predict the way in which the anatomy will/can deform. The second method is related, and it uses the statistical model in conjunction with a biomechanical model of anatomical deformation. It examines the principal modes of co-variation between shape and forces, with the latter driving the biomechanical model, and thus predicting deformation. Results are shown on simulated images, demonstrating that systematic deformations, such as those resulting from change in position or from tumor growth, can be estimated very well using these models. Estimation accuracy will depend on the application, and particularly on how systematic a deformation of interest is.

Finite element modeling of brain tumor mass-effect from 3d medical images

Motivated by the need for methods to aid the deformable registration of brain tumor images, we present a three-dimensional (3D) mechanical model for simulating large non-linear deformations induced by tumors to the surrounding encephalic tissues. The model is initialized with 3D radiological images and is implemented using the finite element (FE) method. To simulate the widely varying behavior of brain tumors, the model is controlled by a number of parameters that are related to variables such as the bulk tumor location, size, mass-effect, and peri-tumor edema extent. Model predictions are compared to real brain tumor-induced deformations observed in serial-time MRI scans of a human subject and 3 canines with surgically transplanted gliomas. Results indicate that the model can reproduce the real deformations with an accuracy that is similar to that of manual placement of landmark points to which the model is compared.

A Combined Statistical and Biomechanical Model for Estimation of Intra-operative Prostate Deformation

An approach for estimating the deformation of the prostate caused by transrectal ultrasound (TRUS) probe insertion is presented. This work is particularly useful during brachytherapy procedures, in which planning for radioactive seed insertion is performed on preoperative scans, and significant deformation of the prostate can occur during the procedure. The approach makes use of a patient specific biomechanical model to run simulations for TRUS probe insertion, extract the main modes of the deformation of the prostate, and use this information to establish a deformable registration between 2 orthogonal cross-sectional ultrasound images and the preoperative prostate. In the work presented here, the approach is tested on an anatomy-realistic biomechanical phantom for the prostate and results are reported for 5 test simulations. More than 73% of maximum deformation of the prostate was recovered, with the estimation error mostly attributed to the relatively small number of biomechanical simulations used for training.

Finite element mesh generation and remeshing from segmented medical images

We present an approach for the automatic generation of patient-specific tetrahedral finite-element (FE) meshes from multiple-label segmented medical images. The approach uses a mesh refinement method with guaranteed tetrahedral element quality and includes a post-processing step with operations to change the mesh topology. Results indicate good approximation of the meshed geometry and acceptable simulation errors for a mechanical problem with a known analytical solution. We also present a method for correcting mesh distortion associated with large-deformation mechanical problems such as those that can arise in dealing with biological soft tissues. This adaptive remeshing scheme is driven by an estimate of the local a posteriori FE error profile. Finally, we demonstrate the use of our approach to carry out a large-deformation simulation of non-infiltrating brain tumor growth starting from a segmented medical image.

Shape Representation via Best Orthogonal Basis Selection

We formulate the problem of finding a statistical representation of shape as a best basis selection problem in which the goal is to choose the basis for optimal shape representation from a very large library of bases. In this work, our emphasis is on applying this basis selection framework using the wavelet packets library to estimate the probability density function of a class of shapes from a limited number of training samples. Wavelet packets offer a large number of complete orthonormal bases which can be searched for the basis that optimally allows the analysis of shape details at different scales. The estimated statistical shape distribution is capable of generalizing to shape examples not encountered during training, while still being specific to the modeled class of shapes. Using contours from two-dimensional MRI images of the corpus callosum, we demonstrate the ability of this approach to approximate the probability distribution of the modeled shapes, even with a few training samples.

A statistical approach for estimating brain tumor induced deformation

A general statistical approach for predicting anatomical deformations is presented. Emphasis in this paper is on estimating deformations induced in the brain anatomy due to tumor growth. The presented approach utilizes the principal modes of co-variation between deformed (after tumor growth) and undeformed (before tumor growth) anatomy to estimate one given the other. In particular, with a statistical model constructed from a number of training samples, a patients brain anatomy prior to tumor growth is estimated based on the patients tumor-bearing images. This approach is suitable for use in registering a patients tumor-bearing images to an anatomical atlas for purposes of surgical, or radio-surgical planning. The proposed approach is tested on a data set of 40 axial 2D brain images of normal human subjects. A biomechanical model was used to simulate tumor growth in each image of the data set. Pairs of deformed and undeformed anatomy were generated by tracking locations of 94 landmark points. The quality of the estimates of the undeformed anatomy are evaluated using the leave-one-out method. Results indicate good estimation accuracy considering the relatively small sample size.

Registration of brain images with tumors: towards the construction of statistical atlases for therapy planning

A deformable registration method is proposed to register a brain atlas with tumor-bearing brain scans. The tumor mass effect is first simulated in the (normal) atlas, using a biomechanical model of mass effect. The tumor-bearing atlas is subsequently warped to the patients scan by a deformable registration method, built upon the idea of HAMMER registration algorithm developed for normal brains. The potential of using the pattern of deformation around the tumor region to optimize the location of tumor seed and other parameters of the tumor model is also explored. Quantitative evaluation on simulated data shows that the proposed method achieves accuracy similar to that achieved in registration of images without tumors. Moreover, limited registration results on real tumors are promising

Deformable registration of brain tumor images via a statistical model of tumor-induced deformation

An approach to the deformable registration of three-dimensional brain tumor images to a normal brain atlas is presented. The approach involves the integration of three components: a biomechanical model of tumor mass-effect, a statistical approach to estimate the models parameters, and a deformable image registration method. Statistical properties of the sought deformation map from the atlas to the image of a tumor patient are first obtained through tumor mass-effect simulations on normal brain images. This map is decomposed into the sum of two components in orthogonal subspaces, one representing inter-individual differences in brain shape, and the other representing tumor-induced deformation. For a new tumor case, a partial observation of the sought deformation map is obtained via deformable image registration and is decomposed into the aforementioned spaces in order to estimate the mass-effect model parameters. Using this estimate, a simulation of tumor mass-effect is performed on the atlas image in order to generate an image that is similar to tumor patients image, thereby facilitating the atlas registration process. Results for a real tumor case and a number of simulated tumor cases indicate significant reduction in the registration error due to the presented approach as compared to the direct use of deformable image registration.