Multi-Disease Segmentation of Gliomas and White Matter Hyperintensities in the BraTS Data Using a 3D Convolutional Neural Network

Abstract

An important challenge in segmenting real-world biomedical imaging data is the presence of multiple disease processes within individual subjects. Most adults above age 60 exhibit a variable degree of small vessel ischemic disease, as well as chronic infarcts, which will manifest as white matter hyperintensities (WMH) on brain MRIs. Subjects diagnosed with gliomas will also typically exhibit some degree of abnormal T2 signal due to WMH, rather than just due to tumor. We sought to develop a fully automated algorithm to distinguish and quantify these distinct disease processes within individual subjects' brain MRIs. To address this multi-disease problem, we trained a 3D U-Net to distinguish between abnormal signal arising from tumors vs. WMH in the 3D multi-parametric MRI (mpMRI, i.e., native T1-weighted, T1-post-contrast, T2, T2-FLAIR) scans of the International Brain Tumor Segmentation (BraTS) 2018 dataset (n training = 285, n validation = 66). Our trained neuroradiologist manually annotated WMH on the BraTS training subjects, finding that 69% of subjects had WMH. Our 3D U-Net model had a 4-channel 3D input patch (80 × 80 × 80) from mpMRI, four encoding and decoding layers, and an output of either four [background, active tumor (AT), necrotic core (NCR), peritumoral edematous/infiltrated tissue (ED)] or five classes (adding WMH as the fifth class). For both the four-and five-class output models, the median Dice for whole tumor (WT) extent (i.e., union of AT, ED, NCR) was 0.92 in both training and validation sets. Notably, the five-class model achieved significantly (p = 0.002) lower/better Hausdorff distances for WT extent in the training subjects. There was strong positive correlation between manually segmented and predicted volumes for WT (r = 0.96) and WMH (r = 0.89). Larger lesion volumes were positively correlated with higher/better Dice scores for WT (r = 0.33), WMH (r = 0.34), and across all lesions (r = 0.89) on a log(10) transformed scale. While the median Dice for WMH was 0.42 across training subjects with WMH, the median Dice was 0.62 for those with

The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification

“…Since the focus of the challenge was on gliomas all other abnormalities (such as white matter hyperintensities that are typically secondary to small vessel ischemic disease) were not considered in the annotation process. This was made particularly apparent from previous efforts that attempted to perform a multidisease segmentation [38].…”

Interinstitutional Portability of a Deep Learning Brain MRI Lesion Segmentation Algorithm

“…One experiment in the study used training data from an additional 198 patients from IN2 with primary brain tumors (World Health Organization grade II-IV gliomas). Another experiment in the study used 285 patients' open-source data from the 2018 Multimodal Brain Tumor Segmentation challenge (BraTS) (9), which also included manual segmentations of white matter hyperintensities, as described in Rudie et al (10).…”

Nonlocal convolutional block attention module VNet for gliomas automatic segmentation

“…As far as we know, some experts and scholars have conducted research on this and achieved considerable results. There are also some novel models proposed, such as 3D‐CNN, 48 cascaded deep CNN, 24 and so on. After learning from previous experience, we proposed a joint model and applied it to the segmentation of 3D brain MR images, and obtained clear results.…”