Traditionally, radiologists have crudely quantified tumor extent by measuring the longest and shortest dimension by dragging a cursor between opposite boundary points across a single image rather than full segmentation of the volumetric extent. For algorithmic-based volumetric segmentation, the degree of radiologist experiential involvement varies from confirming a fully automated segmentation, to making a single drag on an image to initiate semi-automated segmentation, to making multiple drags and clicks on multiple images during interactive segmentation. An experiment was designed to test an algorithm that allows various levels of interaction. Given the ground-truth of the BraTS training data, which delimits the brain tumors of 285 patients on multi-spectral MR, a computer simulation mimicked the process that a radiologist would follow to perform segmentation with real-time interaction. Clicks and drags were placed only where needed in response to the deviation between real-time segmentation results and assumed radiologist's goal, as provided by the ground-truth. Results of accuracy for various levels of interaction are presented along with estimated elapsed time, in order to measure efficiency. Average total elapsed time, including loading the study through confirming 3D contours, was 46 s.

A semi-automatic image segmentation method, called SAMBAS, based on workflow familiar to clinical radiologists is described. The user initializes 3D segmentation by drawing a long axis on a multi-plane reformat (MPR). As the user draws, a 2D segmentation updates in real-time for interactive feedback. When necessary, additional long axes, short axes, or other editing operations may be drawn on one or more MPR planes. The method learns probability distributions from the drawing to perform the MPR segmentation, and in turn, it learns from the MPR segmentation to perform the 3D segmentation. As a preliminary experiment, a batch simulation was performed where long and short axes were automatically drawn on each of 285 multi-spectral MR brain scans of glioma patients in the 2018 BraTS Challenge training data. Average Dice coefficient for tumor core was 0.86, and the Hausdorff-95% distance was 4.4 mm. As another experiment, a convolution neural network was trained on the same data, and applied to the BraTS validation and test data. Its outputs, computed offline, were integrated into the interactive method. Ten volunteers used the interface on the BraTS validation and test data. On the 66 scans of the validation data, average Dice coefficient for core tumor improved from 0.76 with deep learning alone, to 0.82 as an interactive system.

Gliomas are the most common primary brain malignancies, with different degrees of aggressiveness, variable prognosis and various heterogeneous histologic sub-regions, i.e., peritumoral edematous/invaded tissue, necrotic core, active and non-enhancing core. This intrinsic heterogeneity is also portrayed in their radio-phenotype, as their sub-regions are depicted by varying intensity profiles disseminated across multi-parametric magnetic resonance imaging (mpMRI) scans, reflecting varying biological properties. Their heterogeneous shape, extent, and location are some of the factors that make these tumors difficult to resect, and in some cases inoperable. The amount of resected tumor is a factor also considered in longitudinal scans, when evaluating the apparent tumor for potential diagnosis of progression. Furthermore, there is mounting evidence that accurate segmentation of the various tumor sub-regions can offer the basis for quantitative image analysis towards prediction of patient overall survival. This study assesses the state-of-the-art machine learning (ML) methods used for brain tumor image analysis in mpMRI scans, during the last seven instances of the International Brain Tumor Segmentation (BraTS) challenge, i.e., 2012-2018. Specifically, we focus on i) evaluating segmentations of the various glioma sub-regions in pre-operative mpMRI scans, ii) assessing potential tumor progression by virtue of longitudinal growth of tumor sub-regions, beyond use of the RECIST/RANO criteria, and iii) predicting the overall survival from pre-operative mpMRI scans of patients that underwent gross total resection. Finally, we investigate the challenge of identifying the best ML algorithms for each of these tasks, considering that apart from being diverse on each instance of the challenge, the multi-institutional mpMRI BraTS dataset has also been a continuously evolving/growing dataset.