Does Anatomical Contextual Information Improve 3D U-Net-Based Brain Tumor Segmentation?

Abstract

Effective, robust, and automatic tools for brain tumor segmentation are needed for the extraction of information useful in treatment planning. Recently, convolutional neural networks have shown remarkable performance in the identification of tumor regions in magnetic resonance (MR) images. Context-aware artificial intelligence is an emerging concept for the development of deep learning applications for computer-aided medical image analysis. A large portion of the current research is devoted to the development of new network architectures to improve segmentation accuracy by using context-aware mechanisms. In this work, it is investigated whether or not the addition of contextual information from the brain anatomy in the form of white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) masks and probability maps improves U-Net-based brain tumor segmentation. The BraTS2020 dataset was used to train and test two standard 3D U-Net (nnU-Net) models that, in addition to the conventional MR image modalities, used the anatomical contextual information as extra channels in the form of binary masks (CIM) or probability maps (CIP). For comparison, a baseline model (BLM) that only used the conventional MR image modalities was also trained. The impact of adding contextual information was investigated in terms of overall segmentation accuracy, model training time, domain generalization, and compensation for fewer MR modalities available for each subject. Median (mean) Dice scores of 90.2 (81.9), 90.2 (81.9), and 90.0 (82.1) were obtained on the official BraTS2020 validation dataset (125 subjects) for BLM, CIM, and CIP, respectively. Results show that there is no statistically significant difference when comparing Dice scores between the baseline model and the contextual information models (p > 0.05), even when comparing performances for high and low grade tumors independently. In a few low grade cases where improvement was seen, the number of false positives was reduced. Moreover, no improvements were found when considering model training time or domain generalization. Only in the case of compensation for fewer MR modalities available for each subject did the addition of anatomical contextual information significantly improve (p < 0.05) the segmentation of the whole tumor. In conclusion, there is no overall significant improvement in segmentation performance when using anatomical contextual information in the form of either binary WM, GM, and CSF masks or probability maps as extra channels.

MRI brain tumor segmentation using residual Spatial Pyramid Pooling-powered 3D U-Net

“…ASPP has been employed in various studies within brain segmentation. However, as per Tampu et al ( 28 ), boosting context alone does not increase the performance of the model.…”

Automated Brain Tumor Detection and Segmentation for Treatment Response Assessment Using Amino Acid PET

“…These findings are in line with a recent study from Ladefoged et al (25) in which an artificial neural network was developed and trained on 18 F-FET PET and MRI scans from 233 adult brain tumor patients and applied to a dataset of 66 pediatric brain tumor patients for automated tumor segmentation. The authors also found the largest relative errors for tumor segmentations for small tumors with a volume of less than 10 cm 3 . Although the network demonstrated excellent performance in pediatric tumor patients, a few cases were reported in which the network erroneously delineated anatomic regions showing a high physiologic uptake.…”

Selective Deeply Supervised Multi-Scale Attention Network for Brain Tumor Segmentation

“…However, due to their smaller sizes, irregular shapes, and similar textures to the surrounding tissues, enhanced and core tumors remain complex challenges in terms of reliable segmentation [ 2 ]. To date, none of the currently available methods, including [ 2 , 13 , 15 , 16 , 17 , 18 ], have achieved the same level of performance for ET and TC region segmentation (regarding the whole tumor region).…”