Novel Volumetric Sub-region Segmentation in Brain Tumors
A novel deep learning based model called Multi-Planar Spatial Convolutional Neural Network (MPS-CNN) is proposed for effective, automated segmentation of different sub-regions viz. peritumoral edema (ED), necrotic core (NCR), enhancing and non-enhancing tumor core (ET/NET), from multi-modal MR image...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-01-01
|
| Series: | Frontiers in Computational Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fncom.2020.00003/full |
| _version_ | 1811216995677372416 |
|---|---|
| author | Subhashis Banerjee Subhashis Banerjee Sushmita Mitra |
| author_facet | Subhashis Banerjee Subhashis Banerjee Sushmita Mitra |
| author_sort | Subhashis Banerjee |
| collection | DOAJ |
| description | A novel deep learning based model called Multi-Planar Spatial Convolutional Neural Network (MPS-CNN) is proposed for effective, automated segmentation of different sub-regions viz. peritumoral edema (ED), necrotic core (NCR), enhancing and non-enhancing tumor core (ET/NET), from multi-modal MR images of the brain. An encoder-decoder type CNN model is designed for pixel-wise segmentation of the tumor along three anatomical planes (axial, sagittal, and coronal) at the slice level. These are then combined, by incorporating a consensus fusion strategy with a fully connected Conditional Random Field (CRF) based post-refinement, to produce the final volumetric segmentation of the tumor and its constituent sub-regions. Concepts, such as spatial-pooling and unpooling are used to preserve the spatial locations of the edge pixels, for reducing segmentation error around the boundaries. A new aggregated loss function is also developed for effectively handling data imbalance. The MPS-CNN is trained and validated on the recent Multimodal Brain Tumor Segmentation Challenge (BraTS) 2018 dataset. The Dice scores obtained for the validation set for whole tumor (WT :NCR/NE +ET +ED), tumor core (TC:NCR/NET +ET), and enhancing tumor (ET) are 0.90216, 0.87247, and 0.82445. The proposed MPS-CNN is found to perform the best (based on leaderboard scores) for ET and TC segmentation tasks, in terms of both the quantitative measures (viz. Dice and Hausdorff). In case of the WT segmentation it also achieved the second highest accuracy, with a score which was only 1% less than that of the best performing method. |
| first_indexed | 2024-04-12T06:46:36Z |
| format | Article |
| id | doaj.art-21e1350ffb9d4e759f866d56de7da588 |
| institution | Directory Open Access Journal |
| issn | 1662-5188 |
| language | English |
| last_indexed | 2024-04-12T06:46:36Z |
| publishDate | 2020-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Computational Neuroscience |
| spelling | doaj.art-21e1350ffb9d4e759f866d56de7da5882022-12-22T03:43:31ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882020-01-011410.3389/fncom.2020.00003485259Novel Volumetric Sub-region Segmentation in Brain TumorsSubhashis Banerjee0Subhashis Banerjee1Sushmita Mitra2Machine Intelligence Unit, Indian Statistical Institute, Kolkata, IndiaDepartment of CSE, University of Calcutta, Kolkata, IndiaMachine Intelligence Unit, Indian Statistical Institute, Kolkata, IndiaA novel deep learning based model called Multi-Planar Spatial Convolutional Neural Network (MPS-CNN) is proposed for effective, automated segmentation of different sub-regions viz. peritumoral edema (ED), necrotic core (NCR), enhancing and non-enhancing tumor core (ET/NET), from multi-modal MR images of the brain. An encoder-decoder type CNN model is designed for pixel-wise segmentation of the tumor along three anatomical planes (axial, sagittal, and coronal) at the slice level. These are then combined, by incorporating a consensus fusion strategy with a fully connected Conditional Random Field (CRF) based post-refinement, to produce the final volumetric segmentation of the tumor and its constituent sub-regions. Concepts, such as spatial-pooling and unpooling are used to preserve the spatial locations of the edge pixels, for reducing segmentation error around the boundaries. A new aggregated loss function is also developed for effectively handling data imbalance. The MPS-CNN is trained and validated on the recent Multimodal Brain Tumor Segmentation Challenge (BraTS) 2018 dataset. The Dice scores obtained for the validation set for whole tumor (WT :NCR/NE +ET +ED), tumor core (TC:NCR/NET +ET), and enhancing tumor (ET) are 0.90216, 0.87247, and 0.82445. The proposed MPS-CNN is found to perform the best (based on leaderboard scores) for ET and TC segmentation tasks, in terms of both the quantitative measures (viz. Dice and Hausdorff). In case of the WT segmentation it also achieved the second highest accuracy, with a score which was only 1% less than that of the best performing method.https://www.frontiersin.org/article/10.3389/fncom.2020.00003/fullconvolutional neural networkbrain tumor segmentationspatial-pooling and unpoolingconditional random fieldmulti-planar CNNclass imbalance |
| spellingShingle | Subhashis Banerjee Subhashis Banerjee Sushmita Mitra Novel Volumetric Sub-region Segmentation in Brain Tumors Frontiers in Computational Neuroscience convolutional neural network brain tumor segmentation spatial-pooling and unpooling conditional random field multi-planar CNN class imbalance |
| title | Novel Volumetric Sub-region Segmentation in Brain Tumors |
| title_full | Novel Volumetric Sub-region Segmentation in Brain Tumors |
| title_fullStr | Novel Volumetric Sub-region Segmentation in Brain Tumors |
| title_full_unstemmed | Novel Volumetric Sub-region Segmentation in Brain Tumors |
| title_short | Novel Volumetric Sub-region Segmentation in Brain Tumors |
| title_sort | novel volumetric sub region segmentation in brain tumors |
| topic | convolutional neural network brain tumor segmentation spatial-pooling and unpooling conditional random field multi-planar CNN class imbalance |
| url | https://www.frontiersin.org/article/10.3389/fncom.2020.00003/full |
| work_keys_str_mv | AT subhashisbanerjee novelvolumetricsubregionsegmentationinbraintumors AT subhashisbanerjee novelvolumetricsubregionsegmentationinbraintumors AT sushmitamitra novelvolumetricsubregionsegmentationinbraintumors |