3D MRI brain tumor segmentation using autoencoder regularization

By Andriy Myronenko. BrainLes@MICCAI, 2018

1,428 citations recorded by the source corpus.

Automated segmentation of brain tumors from 3D magnetic resonance images (MRIs) is necessary for the diagnosis, monitoring, and treatment planning of the disease. Manual delineation practices require anatomical knowledge, are expensive, time consuming and can be inaccurate due to human error. Here, we describe a semantic segmentation network for tumor subregion segmentation from 3D MRIs based on encoder-decoder architecture. Due to a limited training dataset size, a variational auto-encoder branch is added to reconstruct the input image itself in order to regularize the shared decoder and impose additional constraints on its layers. The current approach won 1st place in the BraTS 2018 challenge.

Subfields: Computer Vision and Pattern Recognition, Neurology

View Paper: https://www.semanticscholar.org/paper/37a18be8c599b781cc28b6a62d8f11e8a6a75169

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Yin Dai
2. Yifan Gao
3. Jun Ma
4. Bo Wang
5. Feifei Li
6. Klaus Hermann Maier-Hein
7. Daguang Xu
8. Fabian Isensee

In Neurology

1. Yin Dai
2. Yifan Gao
3. Daguang Xu
4. Jun Ma
5. Feifei Li
6. Bo Wang
7. Su Ruan
8. Stéphane Canu

Citations in the corpus

• AbdomenCT-1K: Is Abdominal Organ Segmentation a Solved Problem? (2020, background): “A similar phenomenon has also been found by the winner solution [104] in the well-known brain tumor segmentation (BraTS) challenge 2018.”
• Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge (2018, background): “C V ] 2 3 A pr 2 01 9 Moreno Lopez132, Pablo Ribalta Lorenzo72, Zhentai Lu53, Lin Luo31, Zhigang Luo162, Jun Ma73, Kai Ma117, Thomas Mackie60, Anant Madabhushi129, Issam Mahmoudi74, Klaus H. Maier-Hein22, Pradipta Maji36, CP Mammen161, Andreas Mang165, B. S. Manjunath79, Michal Marcinkiewicz71, Stev”
• Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images (2022, methodology): “Recently, deep learning-based brain tumor segmentation techniques [31,21,32,20] have achieved state-of-the-art performance in various benchmarks [7,35,2].”
• U-KAN Makes Strong Backbone for Medical Image Segmentation and Generation (2024, background): “Medical image segmentation is a challenging task to which deep learning methods have been extensively applied and achieved breakthrough advancements in recent years [40, 49, 62, 71, 77].”
• SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation (2024, methodology): “Necessity for CRC-500 Dataset Colorectal cancer (CRC) is the third most common cancer worldwide among men and women, the second leading cause RU-Net [19] 64 MSDenseNet [12] 43 MSD [21] 190 Zhang et al. [7] 388 Our CRC-500 500 of death related to cancer, and the primary cause of death in…”
• nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation (2024, methodology): “The framework is tested by means of three featured architectures ("SegResNet" [28], "DiNTS" [18], "SwinUNETR" [32]).”
• U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation (2024, methodology): “We compared U-Mamba with two CNN-based segmentation networks (nnU-Net [22] and SegResNet [39]) and two Transformer-based networks (UNETR [17] and SwinUNETR [16]), which are widely used in medical image segmentation competitions.”
• UNETR: Transformers for 3D Medical Image Segmentation (2021, background): “Since the advent of deep learning, FCNNs and in particular encoder-decoder architectures [19,13,14,12] have achieved state-of-the-art results in various medical semantic segmentation tasks [1,22,11].”
• TransMed: Transformers Advance Multi-Modal Medical Image Classification (2021, methodology): “[8]–[10] mainly used the input-level fusion, which is the most common fusion method in multi-modal medical image analysis.”
• Swin-UMamba: Mamba-based UNet with ImageNet-based pretraining (2024, neutral): “We select three types of methods as baseline methods for comprehensive evaluation, including CNN-based (nnU-Net [11], SegResNet [21]), transformer-based Table 2.”
• Generalizing Deep Learning for Medical Image Segmentation to Unseen Domains via Deep Stacked Transformation (2020, methodology): “Such a method is a component in the state-of-the-art brain MRI segmentation algorithm [32].”
• Privacy-preserving Federated Brain Tumour Segmentation (2019, methodology): “For the client-side local training, we adapted the stateof-the-art training pipeline originally designed for data-centralised training [7] and implemented as a part of the NVIDIA Clara Train SDK.”