Towards a Computer Aided Prognosis for Brain Glioblastomas Tumor Growth Estimation.

Bridging the gap between mathematical and biological models and clinical applications could be considered as one of the new challenges of medical image analysis over the ten last years. This paper presents an advanced and convivial algorithm for brain glioblastomas tumor growth modelization. The brain glioblastomas tumor region would be extracted using a fast distribution matching developed algorithm based on global pixel wise information. A new model to simulate the tumor growth based on two major elements: cellular automata and fast marching method (CFMM) has been developed and used to estimate the brain tumor evolution during the time. On the basis of this model, experiments were carried out on twenty pathological MRI selected cases that were carefully discussed with the clinical part. The obtained simulated results were validated with ground truth references (real tumor growth measure) using dice metric parameter. As carefully discussed with the clinical partner, experimental results showed that our proposed algorithm for brain glioblastomas tumor growth model proved a good agreement. Our main purpose behind this research was of course to make advances and progress during clinical explorations helping therefore radiologists in their diagnosis. Clinical decisions and guidelines would be hence so more focused with such an advanced tool that could help clinicians and ensuring more accuracy and objectivity.

3D multimodal MRI brain glioma tumor and edema segmentation: a graph cut distribution matching approach.

This study investigates a fast distribution-matching, data-driven algorithm for 3D multimodal MRI brain glioma tumor and edema segmentation in different modalities. We learn non-parametric model distributions which characterize the normal regions in the current data. Then, we state our segmentation problems as the optimization of several cost functions of the same form, each containing two terms: (i) a distribution matching prior, which evaluates a global similarity between distributions, and (ii) a smoothness prior to avoid the occurrence of small, isolated regions in the solution. Obtained following recent bound-relaxation results, the optima of the cost functions yield the complement of the tumor region or edema region in nearly real-time. Based on global rather than pixel wise information, the proposed algorithm does not require an external learning from a large, manually-segmented training set, as is the case of the existing methods. Therefore, the ensuing results are independent of the choice of a training set. Quantitative evaluations over the publicly available training and testing data set from the MICCAI multimodal brain tumor segmentation challenge (BraTS 2012) demonstrated that our algorithm yields a highly competitive performance for complete edema and tumor segmentation, among nine existing competing methods, with an interesting computing execution time (less than 0.5s per image).