Search

Accurate segmentation of lymph nodes in head and neck (H&N) CT images is essential for the radiotherapy planning of the H&N cancer. Atlas-based segmentation methods are widely used for the automated segmentation of such structures. Multi-atlas approaches are proven to be more accurate and robust than using a single atlas. We have recently proposed a general Markov random field (MRF)-based framework that can perform edge-preserving smoothing of the labels at the time of fusing the labels itself. There are three main contributions of this paper: First, we reformulate the "shape based averaging" (SBA) fusion method to fit into the general MRF-based fusion framework. Second, we evaluate the following fusion algorithms for the segmentation of H&N lymph nodes: (i) STAPLE, (ii) SBA, (iii) SBA+MRF, (iv) majority voting (MV), (v) MV+MRF, (vi) global weighted voting (GWV), (vii) GWV+MRF, (viii) local weighted voting (LWV) and (ix) LWV+MRF. Finally, we also study the effect varying the number of atlases on the performance of the above algorithms.

This paper presents 3-D brain tissue classification schemes using three recent promising energy minimization methods for Markov random fields: graph cuts, loopy belief propagation and tree-reweighted message passing. The classification is performed using the well known finite Gaussian mixture Markov Random Field model. Results from the above methods are compared with widely used iterative conditional modes algorithm. The evaluation is performed on a dataset containing simulated T1-weighted MR brain volumes with varying noise and intensity non-uniformities. The comparisons are performed in terms of energies as well as based on ground truth segmentations, using various quantitative metrics.

Validation is the main bottleneck preventing the adoption of many medical image processing algorithms in the clinical practice. In the classical approach, a-posteriori analysis is performed based on some objective metrics. In this work, a different approach based on Petri Nets (PN) is proposed. The basic idea consists in predicting the accuracy that will result from a given processing based on the characterization of the sources of inaccuracy of the system. Here we propose a proof of concept in the scenario of a diffusion imaging analysis pipeline. A PN is built after the detection of the possible sources of inaccuracy. By integrating the first qualitative insights based on the PN with quantitative measures, it is possible to optimize the PN itself, to predict the inaccuracy of the system in a different setting. Results show that the proposed model provides a good prediction performance and suggests the optimal processing approach.

We propose an algorithm for very high-resolution satellite image classification that combines non-supervised segmentation with a supervised classification. Both multi-spectral data and local spatial priors are used in the Gaussian Hidden Markov Random Field (GHMRF) model for the segmentation. Then, two classifiers, Mahalanobis distance classifier and SVM, are studied using intensity, texture and shape features. Validation is done qualitatively and quantitatively by comparison with a manual classification used as a ground truth. Results show very good performance of our approach in comparison to existing techniques. Also, we demonstrate that spectral and spatial features calculated on segmented regions are much more discriminant than the spectral features of the pixels taken individually for the classification task.