Fully Convolutional Neural Network for Improved Brain Segmentation

Khaled, Afifa; Han, Jian-Jun; Ghaleb, Taher Ahmed; Mohamed, Radman

doi:10.1007/s13369-022-07169-7

Cited by 3 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Temporal ensembling [6] aggregates the predictions of multiple network evaluations into an ensemble prediction, and the mean teacher method [7] takes a supervised architecture and makes a copy of the model. In the medical image community, the mean teacher model is good for getting better predictions and has the ability to optimally exploit unlabeled data during training [7,8]. This paper addresses the above limitations and challenges by proposing a 3D-DenseUNet model based on the U-net model, which focuses on loosening information loss and can leverage overall relationships between structures using multi-head attention.…”

Section: Introductionmentioning

confidence: 99%

Two Independent Teachers are Better Role Model

Khaled

Mubarak

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Recent deep learning models have attracted substantial attention in infant brainanalysis. These models have performed state-of-the-art performance, such as semi-supervised techniques (e.g., Temporal Ensembling, mean teacher). However, these models depend on an encoder-decoder structure with stacked local operators togather long-range information, and the local operators limit the efficiency and effectiveness. Besides, the MRI data contain different tissue properties (T P s) suchas T1 and T2. One major limitation of these models is that they use both dataas inputs to the segment process, i.e., the models are trained on the dataset once, and it requires much computational and memory requirements during inference. In this work, we address the above limitations by designing a new deep-learningmodel, called 3D-DenseUNet, which works as adaptable global aggregation blocksin down-sampling to solve the issue of spatial information loss. The self-attention module connects the down-sampling blocks to up-sampling blocks, and integrates the feature maps in three dimensions of spatial and channel, effectively improvingthe representation potential and discriminating ability of the model. Additionally, we propose a new method called Two Independent Teachers (2IT), that summarizes the model weights instead of label predictions. Each teacher model is trainedon different types of brain data, T1 and T2, respectively. Then, a fuse model is added to improve test accuracy and enable training with fewer parameters andlabels compared to the Temporal Ensembling method without modifying the network architecture. Empirical results demonstrate the effectiveness of the proposed method. The code is publicly available at: https://github.com/AfifaKhaled/Two-Independent-Teachers-are-Better-Role-Model

show abstract

Section: Introductionmentioning

confidence: 99%

Two Independent Teachers are Better Role Model

Khaled

Mubarak

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

3D-DenseUNet: A New Model for BrainSegmentation in Magnetic Resonance Images

Khaled,

Mubarak,

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Recent deep learning models have attracted substantial attention in infant brainanalysis. These models have performed state-of-the-art performance, such as semisupervised techniques (e.g., Temporal Ensembling, mean teacher). However, thesemodels depend on an encoder-decoder structure with stacked local operators togather long-range information, and the local operators limit the efficiency and effectiveness. Besides, the MRI data contain different tissue properties (T P s) suchas T1 and T2. One major limitation of these models is that they use both dataas inputs to the segment process, i.e., the models are trained on the dataset once,and it requires much computational and memory requirements during inference. In this work, we address the above limitations by designing a new deep-learningmodel, called 3D-DenseUNet, which works as adaptable global aggregation blocksin down-sampling to solve the issue of spatial information loss. The self-attentionmodule connects the down-sampling blocks to up-sampling blocks, and integrates the feature maps in three dimensions of spatial and channel, effectively improvingthe representation potential and discriminating ability of the model. Additionally,we propose a new method called Two Independent Teachers (2IT), that summarizes the model weights instead of label predictions. Each teacher model is trainedon different types of brain data, T1 and T2, respectively. Then, a fuse model isadded to improve test accuracy and enable training with fewer parameters andlabels compared to the Temporal Ensembling method without modifying the network architecture. Empirical results demonstrate the effectiveness of the proposedmethod. The code is publicly available at: https://github.com/AfifaKhaled/Two-Independent-Teachers-are-Better-Role-Model

show abstract

MRI-GAN: Generative Adversarial Network for Brain Segmentation

Khaled,

Ghaleb

2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Fully Convolutional Neural Network for Improved Brain Segmentation

Cited by 3 publications

References 29 publications

Two Independent Teachers are Better Role Model

Two Independent Teachers are Better Role Model

3D-DenseUNet: A New Model for BrainSegmentation in Magnetic Resonance Images

MRI-GAN: Generative Adversarial Network for Brain Segmentation

Contact Info

Product

Resources

About