1,442 research outputs found
Mathematical preliminaries
No full textIn this chapter we review well-known mathematical tools that will be used throughout the book. Specifically, we review basic concepts on images and measures of their quality, vector and matrix results, linear processing in the image domain and its associated Fourier domain, calculus and variations, and we finish with some notions on shape analysis. Some results are analytically derived so that the reader can gain insight in some topics considered more relevant.</p
Region-Enhanced Joint Dictionary Learning for Cross-Modality Synthesis in Diffusion Tensor Imaging
No full textDiffusion tensor imaging (DTI) has notoriously long acquisition times, and the sensitivity of the tensor computation often make this technique vulnerable to various interferences, for example, physiological motions, limited scanning time and patients with different medical conditions. In neuroimaging, studies usually involve different modalities. We considered the problem of inferring key information in DTI from other modalities. To address such a problem, several cross-modality image synthesis approaches have been proposed recently, in which the content of an image modality is reproduced based on those of another modality. However, these methods typically focus on two modalities of same complexity. In this work we propose a region-enhanced joint dictionary learning method that combines the region-specific information in a joint learning manner. The proposed method encodes intrinsic differences among different modalities, while the jointly learned dictionaries preserve common structures among them. Experimental results show that our approach has desirable properties on cross-modality image synthesis in diffusion tensor images.</p
Geometry Regularized Joint Dictionary Learning for Cross-Modality Image Synthesis in Magnetic Resonance Imaging
No full textMulti-sequence MRI protocols are used in comprehensive examinations of various pathologies in both clinical diagnosis and medical research. Various MRI techniques provide complementary information about living tissue. However, a comprehensive examination covering all modalities is rarely achieved due to considerations of cost, patient comfort, and scanner time availability. This may lead to incomplete records owing to image artifacts or corrupted or lost data. In this paper, we explore the problem of synthesizing images for one MRI modality from an image of another MRI modality of the same subject using a novel geometry regularized joint dictionary learning framework for non-local patch reconstruction. Firstly, we learn a cross-modality joint dictionary from a multi-modality image database. Training image pairs are first co-registered. A cross-modality dictionary pair is then jointly learned by minimizing the cross-modality divergence via a Maximum Mean Discrepancy term in the objective function of the learning scheme. This guarantees that the distribution of both image modalities is taken jointly into account when building the resulting sparse representation. In addition, in order to preserve intrinsic geometrical structure of the synthesized image patches, we further introduced a graph Laplacian regularization term into the objective function. Finally, we present a patch-based non-local reconstruction scheme, providing further fidelity of the synthesized images. Experimental results demonstrate that our method achieves significant performance gains over previously published techniques
Preface to the Proceedings of Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015
No full text
An atlas- and data-driven approach to initializing reaction-diffusion systems in computer cardiac electrophysiology
No full textThe cardiac electrophysiology (EP) problem is governed by a nonlinear anisotropic reaction-diffusion system with a very rapidly varying reaction term associated with the transmembrane cell current. The nonlinearity associated with the cell models requires a stabilization process before any simulation is performed. More importantly, when used in a 3-dimensional (3D) anatomy, it is not sufficient to perform this stabilization on the basis of isolated cells only, since the coupling of the different cells through the tissue greatly modulates the dynamics of the system. Therefore, stabilization of the system must be performed on the entire 3D model. This work develops a novel procedure for the initialization of reaction-diffusion systems for numerical simulations of cardiac EP from steady-state conditions. We exploit surface point correspondence to establish volumetric point correspondence. Upon introduction of a new 3D anatomy with surface point correspondence, a prediction of the cell model steady states is derived from the set of earlier biophysical simulations. We show that the prediction error is typically less than 10% for all model variables, with most variables showing even greater accuracy. When initializing simulations with the predicted model states, it is demonstrated that simulation times can be cut by at least two-thirds and potentially more, which saves hours or days of high-performance computing. Overall, these results increase the clinical applicability of detailed computational EP studies on personalized anatomies
Simulation and synthesis in medical imaging:Second international workshop, SASHIMI 2017 held in conjunction with MICCAI 2017 Québec city, QC, Canada, September 10, 2017 proceedings
No full textFUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare
No full textSimulation and synthesis in medical imaging:Second international workshop, SASHIMI 2017 held in conjunction with MICCAI 2017 Québec city, QC, Canada, September 10, 2017 proceedings
No full text- …
