ISLES 2016 and 2017-Benchmarking Ischemic Stroke Lesion Outcome Prediction Based on Multispectral MRI

Abstract

Performance of models highly depend not only on the used algorithm but also the data set it was applied to. This makes the comparison of newly developed tools to previously published approaches difficult. Either researchers need to implement others' algorithms first, to establish an adequate benchmark on their data, or a direct comparison of new and old techniques is infeasible. The Ischemic Stroke Lesion Segmentation (ISLES) challenge, which has ran now consecutively for 3 years, aims to address this problem of comparability. ISLES 2016 and 2017 focused on lesion outcome prediction after ischemic stroke: By providing a uniformly pre-processed data set, researchers from all over the world could apply their algorithm directly. A total of nine teams participated in ISLES 2015, and 15 teams participated in ISLES 2016. Their performance was evaluated in a fair and transparent way to identify the state-of-the-art among all submissions. Top ranked teams almost always employed deep learning tools, which were predominately convolutional neural networks (CNNs). Despite the great efforts, lesion outcome prediction persists challenging. The annotated data set remains publicly available and new approaches can be compared directly via the online evaluation system, serving as a continuing benchmark (www.isles-challenge.org).

Prediction of final infarct volume from native CT perfusion and treatment parameters using deep learning

“…This is not an image segmentation task, where all the relevant information is contained in the input, but a prediction problem with a ground truth that is defined on images acquired several days later. For example, the ISLES 2017 challenge (Winzeck et al, 2018) has a similar goal, namely predicting the final infarct from acute DWI and MR perfusion imaging, and reported a mean Dice score of 0.32 the best result. Of course, a direct comparison between these values is not possible since the ISLES challenge had different modalities (with especially the DWI imaging being very informative compared to perfusion imaging) and a different population (mostly early successful early recanalization, which results in small lesions and hence lower Dice scores), but it illustrates the difficulty of the problem.…”

Ischemic Stroke Lesion Segmentation in CT Perfusion Scans Using Pyramid Pooling and Focal Loss

“…The ISLES challenge was established to fairly compare approaches in stroke lesion segmentation and characterization [17,25], resulting in the development of effective CNN models for this task [4,10,13]. All previous challenges focused on multispectral MRI data [17,25]. Many early approaches focused on analyzing patches, in part due to memory issues.…”

“…Treatment of stoke is time sensitive, requiring tissue reperfusion within less than 4-6 hours of stroke onset [17]. Current standards for evaluating stroke requires manual segmentation in MRI or CT images [1,9,17], a challenging and time consuming task, due to the changing appearance of lesions over time and their presence in various locations in the brain [17,25]. There is a growing need for automatic segmentation methods to accurately identify lesions and to help plan effective treatment.…”

Exploiting Bilateral Symmetry in Brain Lesion Segmentation with Reflective Registration

“…While a plethora of automatic lesion segmentation methods has been proposed, most of the currently leading methods are based on convolutional neural networks (CNN) [15]. Many of these use 2D CNNs, where the 3D neuroimage is processed as a sequence of independent 2D slices.…”