Optimized U-Net for Brain Tumor Segmentation
- URL: http://arxiv.org/abs/2110.03352v1
- Date: Thu, 7 Oct 2021 11:44:09 GMT
- Title: Optimized U-Net for Brain Tumor Segmentation
- Authors: Micha{\l} Futrega, Alexandre Milesi, Michal Marcinkiewicz, Pablo
Ribalta
- Abstract summary: We propose an optimized U-Net architecture for a brain mboxtumor segmentation task in the BraTS21 Challenge.
Our solution was the winner of the challenge validation phase, with the normalized statistical ranking score of 0.267 and mean Dice score of 0.8855.
- Score: 62.997667081978825
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: We propose an optimized U-Net architecture for a brain \mbox{tumor}
segmentation task in the BraTS21 Challenge. To find the \mbox{optimal} model
architecture and learning schedule we ran an extensive ablation study to test:
deep supervision loss, Focal loss, decoder attention, drop block, and residual
connections. Additionally, we have searched for the optimal depth of the U-Net
and number of convolutional channels. Our solution was the winner of the
challenge validation phase, with the normalized statistical ranking score of
0.267 and mean Dice score of 0.8855
Related papers
- Searching for Effective Neural Network Architectures for Heart Murmur
Detection from Phonocardiogram [5.183688633606942]
The George B. Moody PhysioNet Challenge 2022 raised problems of heart murmur detection and related abnormal cardiac function identification from phonocardiograms (PCGs)
This work describes the novel approaches developed by our team, Revenger, to solve these problems.
arXiv Detail & Related papers (2023-03-06T09:31:42Z) - Learning To Dive In Branch And Bound [95.13209326119153]
We propose L2Dive to learn specific diving structurals with graph neural networks.
We train generative models to predict variable assignments and leverage the duality of linear programs to make diving decisions.
arXiv Detail & Related papers (2023-01-24T12:01:45Z) - Extending nnU-Net is all you need [2.1729722043371016]
We use nnU-Net to participate in the AMOS2022 challenge, which comes with a unique set of tasks.
The dataset is one of the largest ever created and boasts 15 target structures.
Our final ensemble achieves Dice scores of 90.13 for Task 1 (CT) and 89.06 for Task 2 (CT+MRI) in a 5-fold cross-validation.
arXiv Detail & Related papers (2022-08-23T07:54:29Z) - DOGE-Train: Discrete Optimization on GPU with End-to-end Training [28.795080637690095]
We present a fast, scalable, data-driven approach for solving relaxations of 0-1 integer linear programs.
We use a combination of graph neural networks (GNN) and the Lagrange decomposition based algorithm FastDOG.
arXiv Detail & Related papers (2022-05-23T21:09:41Z) - HNF-Netv2 for Brain Tumor Segmentation using multi-modal MR Imaging [86.52489226518955]
We extend our HNF-Net to HNF-Netv2 by adding inter-scale and intra-scale semantic discrimination enhancing blocks.
Our method won the RSNA 2021 Brain Tumor AI Challenge Prize (Segmentation Task)
arXiv Detail & Related papers (2022-02-10T06:34:32Z) - Generalized Wasserstein Dice Loss, Test-time Augmentation, and
Transformers for the BraTS 2021 challenge [3.3180658085204513]
Brain tumor segmentation is a challenging task in medical image computation.
In this paper, we explore strategies to increase model robustness without increasing inference time.
Our ensemble of seven 3D U-Nets with test-time augmentation produces an average dice score of 89.4% and an average Hausdorff 95% distance of 10.0 mm.
arXiv Detail & Related papers (2021-12-24T13:01:44Z) - H2NF-Net for Brain Tumor Segmentation using Multimodal MR Imaging: 2nd
Place Solution to BraTS Challenge 2020 Segmentation Task [96.49879910148854]
Our H2NF-Net uses the single and cascaded HNF-Nets to segment different brain tumor sub-regions.
We trained and evaluated our model on the Multimodal Brain Tumor Challenge (BraTS) 2020 dataset.
Our method won the second place in the BraTS 2020 challenge segmentation task out of nearly 80 participants.
arXiv Detail & Related papers (2020-12-30T20:44:55Z) - Generalized Wasserstein Dice Score, Distributionally Robust Deep
Learning, and Ranger for brain tumor segmentation: BraTS 2020 challenge [4.787559241352888]
Training a deep neural network is an optimization problem with four main ingredients: the design of the deep neural network, the per-sample loss function, the population loss function, and the architectures.
We experimented with adopting the opposite approach, focusing only on the design of the deep neural network.
With an ensemble of three deep neural networks trained with various optimization procedures, we achieved promising results on the validation of the BraTS 2020 challenge.
arXiv Detail & Related papers (2020-11-03T10:50:48Z) - Brain tumor segmentation with self-ensembled, deeply-supervised 3D U-net
neural networks: a BraTS 2020 challenge solution [56.17099252139182]
We automate and standardize the task of brain tumor segmentation with U-net like neural networks.
Two independent ensembles of models were trained, and each produced a brain tumor segmentation map.
Our solution achieved a Dice of 0.79, 0.89 and 0.84, as well as Hausdorff 95% of 20.4, 6.7 and 19.5mm on the final test dataset.
arXiv Detail & Related papers (2020-10-30T14:36:10Z) - Efficient Optimistic Exploration in Linear-Quadratic Regulators via
Lagrangian Relaxation [107.06364966905821]
We study the exploration-exploitation dilemma in the linear quadratic regulator (LQR) setting.
Inspired by the extended value iteration algorithm used in optimistic algorithms for finite MDPs, we propose to relax the optimistic optimization of ofulq.
We show that an $epsilon$-optimistic controller can be computed efficiently by solving at most $Obig(log (1/epsilon)big)$ Riccati equations.
arXiv Detail & Related papers (2020-07-13T16:30:47Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.