Improving the Generalizability of Convolutional Neural Network-Based Segmentation on CMR Images

@article{Chen2020ImprovingTG,
  title={Improving the Generalizability of Convolutional Neural Network-Based Segmentation on CMR Images},
  author={Chen Chen and Wenjia Bai and Rhodri H Davies and Anish N Bhuva and Charlotte H. Manisty and James C. Moon and Nay Linn Aung and Aaron M. Lee and Mihir M. Sanghvi and Kenneth Fung and Jos{\'e} Miguel Paiva and Steffen Erhard Petersen and Elena Lukaschuk and Stefan K. Piechnik and Stefan Neubauer and Daniel Rueckert},
  journal={Frontiers in Cardiovascular Medicine},
  year={2020},
  volume={7}
}
Background: Convolutional neural network (CNN) based segmentation methods provide an efficient and automated way for clinicians to assess the structure and function of the heart in cardiac MR images. While CNNs can generally perform the segmentation tasks with high accuracy when training and test images come from the same domain (e.g., same scanner or site), their performance often degrades dramatically on images from different scanners or clinical sites. Methods: We propose a simple yet… 
A Generalizable Deep-Learning Approach for Cardiac Magnetic Resonance Image Segmentation Using Image Augmentation and Attention U-Net
TLDR
A novel combination of image augmentation and a consistency loss function to improve model robustness to typical variations in CMRI data and demonstrated improved performance on CMRI images from new vendors and centers.
Studying Robustness of Semantic Segmentation under Domain Shift in cardiac MRI
TLDR
This work builds upon a fixed U-Net architecture configured by the nnU-net framework to investigate various data augmentation techniques and batch normalization layers as an easy-to-customize pipeline component and provide general guidelines on how to improve domain generalizability abilities in existing deep learning methods.
Cooperative Training and Latent Space Data Augmentation for Robust Medical Image Segmentation
TLDR
A cooperative framework for training image segmentation models and a latent space augmentation method for generating hard examples that improves model generalization and robustness with limited data is presented.
Improving accuracy and robustness of deep convolutional neural network based thoracic OAR segmentation.
TLDR
It was found that even a subtle shift of organs inside patient body, caused by abdominal compression technique used in the authors' institution, can confuse a well-trained DCNN, so incorporating cases from local institution into training process improves the accuracy and robustness of the DCNN model.
Graph Cuts Loss to Boost Model Accuracy and Generalizability for Medical Image Segmentation
  • Zhou Zheng, M. Oda, K. Mori
  • Computer Science
    2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
  • 2021
TLDR
A novel, robust yet straightforward loss function to boost model accuracy and generalizability for medical image segmentation, reformulate the graph cuts cost function to a loss function for supervised learning and extends it to other segmentation tasks.
Improving the generalization capability of deep learning-based algorithms for ventricular cavity segmentation using simulated CMR images
The generalization capability of deep learning-based segmentation algorithms across different sites and vendors, as well as MRI data with high variance in contrast, is limited. This affects the
DoFE: Domain-Oriented Feature Embedding for Generalizable Fundus Image Segmentation on Unseen Datasets
TLDR
A novel Domain-oriented Feature Embedding (DoFE) framework to improve the generalization ability of CNNs on unseen target domains by exploring the knowledge from multiple source domains and generates satisfying segmentation results on unseen datasets and exceeds other domain generalization and network regularization methods.
Deep Learning for Cardiac Image Segmentation: A Review
TLDR
A review of over 100 cardiac image segmentation papers using deep learning, which covers common imaging modalities including magnetic resonance imaging, computed tomography, and ultrasound and major anatomical structures of interest.
Domain-Adversarial Learning for Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac MR Image Segmentation
TLDR
This work proposes a principled solution to the problem of domain shift in CMR by using domain-adversarial learning to train a domain-invariant 2D U-Net using labelled and unlabelled data and shows improved performance as compared to standard training.
Automatic Retinal Blood Vessel Segmentation Based on Fully Convolutional Neural Networks
TLDR
This paper proposes an automatic retinal vessel segmentation framework using deep fully convolutional neural networks (FCN), which integrate novel methods of data preprocessing, data augmentation, and full convolutionAL neural networks that automatically and efficiently performs retinal vessels segmentation.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 67 REFERENCES
A Fully Convolutional Neural Network for Cardiac Segmentation in Short-Axis MRI
TLDR
This work proposes to tackle the problem of automated left and right ventricle segmentation through the application of a deep fully convolutional neural network architecture for pixel-wise labeling in cardiac magnetic resonance imaging.
U-Net: Convolutional Networks for Biomedical Image Segmentation
TLDR
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.
Automated cardiovascular magnetic resonance image analysis with fully convolutional networks
TLDR
An automated analysis method based on a fully convolutional network achieves a performance on par with human experts in analysing CMR images and deriving clinically relevant measures.
An Exploration of 2D and 3D Deep Learning Techniques for Cardiac MR Image Segmentation
TLDR
A fully automated framework for segmentation of the left and right ventricular cavities and the myocardium on short-axis cardiac MR images is presented and it is found that processing the images in a slice-by-slice fashion using 2D networks is beneficial due to a relatively large slice thickness.
3-D Consistent and Robust Segmentation of Cardiac Images by Deep Learning With Spatial Propagation
TLDR
A method based on deep learning to perform cardiac segmentation on short axis Magnetic resonance imaging stacks iteratively from the top slice to the bottom slice iteratively using a novel variant of the U-net.
Fully convolutional multi‐scale residual DenseNets for cardiac segmentation and automated cardiac diagnosis using ensemble of classifiers
TLDR
The approach combined both cardiac segmentation and disease diagnosis into a fully automated framework which is computationally efficient and hence has the potential to be incorporated in computer‐aided diagnosis (CAD) tools for clinical application.
Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?
TLDR
How far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies is measured, to open the door to highly accurate and fully automatic analysis of cardiac CMRI.
Cardiac MRI Segmentation with Strong Anatomical Guarantees
TLDR
With this method, CNNs can now produce results that are both within the inter-expert variability and always anatomically plausible, and can accommodate any cardiac segmentation method and convert its anatomical implausible results to plausible ones without affecting its overall geometric and clinical metrics.
Deep Learning-based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study.
TLDR
A deep learning-based method trained on a data set with high variability can achieve fully automated and accurate cine MRI analysis on multivendor, multicenter cine MR data.
Obesity May Be Bad: Compressed Convolutional Networks for Biomedical Image Segmentation
Compressed Highly parameterize convolutional neural network (CNN) is essential for practical biomedical image segmentation applications. However, the accuracy of compressed CNN based on current
...
1
2
3
4
5
...