Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation

@article{Panfilov2019ImprovingRO,
  title={Improving Robustness of Deep Learning Based Knee MRI Segmentation: Mixup and Adversarial Domain Adaptation},
  author={Egor Panfilov and Aleksei Tiulpin and Stefan Klein and Miika T. Nieminen and Simo S Saarakkala},
  journal={2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)},
  year={2019},
  pages={450-459}
}
Degeneration of articular cartilage (AC) is actively studied in knee osteoarthritis (OA) research via magnetic resonance imaging (MRI). Segmentation of AC tissues from MRI data is an essential step in quantification of their damage. Deep learning (DL) based methods have shown potential in this realm and are the current state-of-the-art, however, their robustness to heterogeneity of MRI acquisition settings remains an open problem. In this study, we investigated two modern regularization… Expand
Ensemble learning for robust knee cartilage segmentation: data from the osteoarthritis initiative
TLDR
Combining a 2-label U-Net (cartilage vs background) with a multi-class U- net for segmentation of cartilage compartment boosts the accuracy of the deep learning model leading to the recovery of missing annotations in the manual dataset. Expand
Multipath CNN with alpha matte inference for knee tissue segmentation from MRI
TLDR
A novel multipath CNN-based method is proposed, which consists of an encoder– decoder-based segmentation network in combination with a low rank tensor-reconstructed segmentationNetwork, and trimap generation, which effectively utilizes superimposed regions. Expand
Deep Learning-based Segmentation of Knee MRI for Fully Automatic Sub-Regional Morphological Assessment of Cartilage Tissues: Data from the Osteoarthritis Initiative.
TLDR
This work introduces a fully automatic method for segmentation and sub-regional assessment of articular cartilage, and evaluates its predictive power in context of radiographic osteoarthritis progression, and finds most of the significant associations identified using the manual segmentation method. Expand
From classical to deep learning: review on cartilage and bone segmentation techniques in knee osteoarthritis research
TLDR
This is the first review that covers both bone and cartilage segmentation models in recognition that knee osteoarthritis is a “whole joint” disease, as well as highlights on diagnostic values of deep learning in emerging knee osteOarthritis research. Expand
Deep learning in magnetic resonance prostate segmentation: A review and a new perspective
TLDR
The state-of-the-art deep learning algorithms in MR prostate segmentation are reviewed; insights to the field are provided by discussing their limitations and strengths; and an optimised 2D U-Net for MR prostate segmentsation is proposed. Expand
Medical Image Segmentation With Limited Supervision: A Review of Deep Network Models
TLDR
A systematic and up-to-date review of the solutions above, with summaries and comments about the methodologies, and highlights several problems in this field. Expand
Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation
TLDR
This article provides a detailed review of the solutions above, summarizing both the technical novelties and empirical results, and compares the benefits and requirements of the surveyed methodologies and provides recommended solutions. Expand
Enhancing MR Image Segmentation with Realistic Adversarial Data Augmentation
TLDR
Results show that the proposed adversarial data augmentation approach can alleviate the need for labeled data while improving model generalization ability, indicating its practical value in medical imaging applications. Expand
Deep learning for knee osteoarthritis diagnosis and progression prediction from plain radiographs and clinical data
TLDR
The main conclusion is that DL yields excellent performance in the diagnostics of OA and in the prediction of its progression, which exceeded the state-of-the-art or formed completely new benchmarks for the evaluation of diagnostic and predictive model performance in OA. Expand
Automatic knee cartilage and bone segmentation using multi-stage convolutional neural networks: data from the osteoarthritis initiative
  • A. Gatti, M. Maly
  • Medicine
  • Magnetic Resonance Materials in Physics, Biology and Medicine
  • 2021
TLDR
A multi-stage convolutional neural network framework learns to automatically segment knee cartilage tissues and bones from MR images acquired with two sequences, producing efficient, accurate quantifications at varying disease severities. Expand
...
1
2
3
...

References

SHOWING 1-10 OF 50 REFERENCES
Deep-Learning for Tidemark Segmentation in Human Osteochondral Tissues Imaged with Micro-computed Tomography
TLDR
The first application of Deep Learning to PTA-stained osteochondral samples that allows to perform tidemark segmentation in a fully-automatic manner is presented, based on U-Net trained using a combination of binary cross-entropy and soft Jaccard loss. Expand
Automatic Knee Osteoarthritis Diagnosis from Plain Radiographs: A Deep Learning-Based Approach
TLDR
A new transparent computer-aided diagnosis method based on the Deep Siamese Convolutional Neural Network to automatically score knee OA severity according to the Kellgren-Lawrence grading scale is presented. Expand
Automatic Grading of Individual Knee Osteoarthritis Features in Plain Radiographs using Deep Convolutional Neural Networks
TLDR
A robust, automatic method to simultaneously predict KL and OARSI grades in knee radiographs, based on Deep Learning and leverages an ensemble of deep residual networks with 50 layers, squeeze-excitation and ResNeXt blocks is developed. Expand
Unsupervised Cross-Modality Domain Adaptation of ConvNets for Biomedical Image Segmentations with Adversarial Loss
TLDR
This paper proposes an unsupervised domain adaptation framework with adversarial learning for cross-modality biomedical image segmentations, based on a dilated fully convolutional network for pixel-wise prediction. Expand
Use of 2D U-Net Convolutional Neural Networks for Automated Cartilage and Meniscus Segmentation of Knee MR Imaging Data to Determine Relaxometry and Morphometry.
TLDR
U-Net demonstrates efficacy and precision in quickly generating accurate segmentations that can be used to extract relaxation times and morphologic characterization and values that can been used in the monitoring and diagnosis of OA. Expand
Automatic segmentation of high- and low-field knee MRIs using knee image quantification with data from the osteoarthritis initiative
TLDR
A segmentation framework for fully automatic segmentation of knee MRI that combines multiatlas rigid registration with voxel classification and was trained on manual segmentations with varying configurations of bones, cartilages, and menisci demonstrated precision and accuracy comparable to manual segmentation. Expand
Knee menisci segmentation using convolutional neural networks: data from the Osteoarthritis Initiative.
TLDR
It was confirmed that medial meniscal extrusion is a predictor for incident OA, and especially for medial menisci an excellent segmentation accuracy was achieved. Expand
A Lifelong Learning Approach to Brain MR Segmentation Across Scanners and Protocols
TLDR
This work proposes a single CNN with shared convolutional filters and domain-specific batch normalization layers, which can be tuned to new domains with only a few (\({\approx 4) labelled images), which largely closes the gap to the benchmark, which is training a dedicated CNN for each scanner. Expand
Semantic-Aware Generative Adversarial Nets for Unsupervised Domain Adaptation in Chest X-ray Segmentation
TLDR
A novel unsupervised domain adaptation approach for segmentation tasks by designing semantic-aware generative adversarial networks (GANs) that transform the test image into the appearance of source domain, with the semantic structural information being well preserved. Expand
Multimodal Machine Learning-based Knee Osteoarthritis Progression Prediction from Plain Radiographs and Clinical Data
TLDR
A multi-modal machine learning-based OA progression prediction model that utilises raw radiographic data, clinical examination results and previous medical history of the patient could significantly improve the subject selection process for OA drug-development trials and help the development of personalised therapeutic plans. Expand
...
1
2
3
4
5
...