Academic Journal
Contextual information extraction in brain tumour segmentation
| Title: | Contextual information extraction in brain tumour segmentation |
|---|---|
| Authors: | Maryam Zia, Usman Ali Baig, Zaka Ur Rehman, Muhammad Yaqub, Shahzad Ahmed, Yudong Zhang, Shuihua Wang, Rizwan Khan |
| Source: | IET Image Processing, Vol 17, Iss 12, Pp 3371-3391 (2023) |
| Publisher Information: | Institution of Engineering and Technology (IET), 2023. |
| Publication Year: | 2023 |
| Subject Terms: | Artificial intelligence, MRI Segmentation, Brain Tumors, Geometry, Classification of Brain Tumor Type and Grade, attention gate, 02 engineering and technology, Image Segmentation, Pattern recognition (psychology), Feature Extraction, Dropout (neural networks), QA76.75-76.765, 03 medical and health sciences, 0302 clinical medicine, Segmentation, Context (archaeology), convolutional neural networks, Machine learning, Photography, 0202 electrical engineering, electronic engineering, information engineering, FOS: Mathematics, Image Segmentation Techniques, Computer software, modified 3D U‐net, TR1-1050, 10. No inequality, Biology, Voxel, Life Sciences, Paleontology, context aware 3D ARDUNet, Computer science, attentional residual dropout block, 3. Good health, Algorithm, Neurology, residual dropout block, Residual, Computer Science, Physical Sciences, Deep Learning in Computer Vision and Image Recognition, Semantic Segmentation, Computer vision, Computer Vision and Pattern Recognition, Block (permutation group theory), Mathematics, Neuroscience |
| Description: | Automatic brain tumour segmentation in MRI scans aims to separate the brain tumour's endoscopic core, edema, non‐enhancing tumour core, peritumoral edema, and enhancing tumour core from three‐dimensional MR voxels. Due to the wide range of brain tumour intensity, shape, location, and size, it is challenging to segment these regions automatically. UNet is the prime three‐dimensional CNN network performance source for medical imaging applications like brain tumour segmentation. This research proposes a context aware 3D ARDUNet (Attentional Residual Dropout UNet) network, a modified version of UNet to take advantage of the ResNet and soft attention. A novel residual dropout block (RDB) is implemented in the analytical encoder path to replace traditional UNet convolutional blocks to extract more contextual information. A unique Attentional Residual Dropout Block (ARDB) in the decoder path utilizes skip connections and attention gates to retrieve local and global contextual information. The attention gate enabled the Network to focus on the relevant part of the input image and suppress irrelevant details. Finally, the proposed Network assessed BRATS2018, BRATS2019, and BRATS2020 to some best‐in‐class segmentation approaches. The proposed Network achieved dice scores of 0.90, 0.92, and 0.93 for the whole tumour. On BRATS2018, BRATS2019, and BRATS2020, tumour core is 0.90, 0.92, 0.93, and enhancing tumour is 0.92, 0.93, 0.94. |
| Document Type: | Article Other literature type |
| Language: | English |
| ISSN: | 1751-9667 1751-9659 |
| DOI: | 10.1049/ipr2.12869 |
| DOI: | 10.60692/zcmke-bqj12 |
| DOI: | 10.60692/2s0b0-n2p39 |
| Access URL: | https://doaj.org/article/9e805cad93cd42bda299c7f9594c4ed2 |
| Rights: | CC BY |
| Accession Number: | edsair.doi.dedup.....d32b536a109e1544651b51e9a9831af9 |
| Database: | OpenAIRE |
Be the first to leave a comment!