Diffuse low-grade gliomas (DLGG) are brain tumors of young adults. They affect the quality of life of the inflicted patients and, if untreated, they evolve into higher grade tumors where the patient's life is at risk. Therapeutic management of DLGGs includes chemotherapy, and tumor diameter is particularly important for the follow-up of DLGG evolution. In fact, the main clinical basis for deciding whether to continue chemotherapy is tumor diameter growth rate. In order to reliably assist the doctors in selecting the most appropriate time to stop treatment, we propose a novel clinical decision support system. Based on two mathematical models, one linear and one exponential, we are able to predict the evolution of tumor diameter under Temozolomide chemotherapy as a first treatment and thus offer a prognosis on when to end it. We present the results of an implementation of these models on a database of 42 patients from Nancy and Montpellier University Hospitals. In this database, 38 patients followed the linear model and four patients followed the exponential model. From a training data set of a minimal size of five, we are able to predict the next tumor diameter with high accuracy. Thanks to the corresponding prediction interval, it is possible to check if the new observation corresponds to the predicted diameter. If the observed diameter is within the prediction interval, the clinician is notified that the trend is within a normal range. Otherwise, the practitioner is alerted of a significant change in tumor diameter.

Graph-based methods have become popular in recent years and have successfully addressed tasks like segmentation and deformable registration. Their main strength is optimality of the obtained solution while their main limitation is the lack of precision due to the grid-like representations and the discrete nature of the quantized search space. In this paper we introduce a novel approach for combined segmentation/registration of brain tumors that adapts graph and sampling resolution according to the image content. To this end we estimate the segmentation and registration marginals towards adaptive graph resolution and intelligent definition of the search space. This information is considered in a hierarchical framework where uncertainties are propagated in a natural manner. State of the art results in the joint segmentation/registration of brain images with low-grade gliomas demonstrate the potential of our approach.

In this paper we propose a novel approach for detection, segmentation and characterization of brain tumors. Our method exploits prior knowledge in the form of a sparse graph representing the expected spatial positions of tumor classes. Such information is coupled with image-based classification techniques along with spatial smoothness constraints towards producing a reliable detection map within a unified graphical model formulation. Towards optimal use of prior knowledge, a two layer interconnected graph is considered with one layer corresponding to the low-grade glioma type (characterization) and the second layer to voxel-based decisions of tumor presence. Efficient linear programming both in terms of performance as well as in terms of computational load is considered to recover the lowest potential of the objective function. The outcome of the method refers to both tumor segmentation as well as their characterization. Promising results on substantial data sets demonstrate the extreme potentials of our method.