A Self-Adaptive Network for Multiple Sclerosis Lesion Segmentation From Multi-Contrast MRI With Various Imaging Sequences

@article{Feng2019ASN,
  title={A Self-Adaptive Network for Multiple Sclerosis Lesion Segmentation From Multi-Contrast MRI With Various Imaging Sequences},
  author={Yushan Feng and Huitong Pan and Craig H. Meyer and Xue Feng},
  journal={2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)},
  year={2019},
  pages={472-475}
}
  • Yushan Feng, Huitong Pan, Xue Feng
  • Published 19 November 2018
  • Computer Science
  • 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)
Deep neural networks have shown promises in the lesion segmentation of multiple sclerosis (MS) from multi-contrast MRI including T1, T2, PD and FLAIR sequences. However, one challenge in deploying such networks into clinical practice is missing MRI sequences due to the variability of image acquisition protocols. Therefore, trained networks need to adapt to practical situations where specific MRI sequences are unavailable. In this paper, we propose a DNN-based MS lesion segmentation framework… 

Figures and Tables from this paper

ModDrop++: A Dynamic Filter Network with Intra-subject Co-training for Multiple Sclerosis Lesion Segmentation with Missing Modalities
TLDR
A novel method to train a unified network adaptive to an arbitrary number of input MRI sequences and upgrades the main idea of ModDrop in two key ways, including a plug-and-play dynamic head and a scaling strategy to improve the expressiveness of the network.
Multiple Sclerosis Lesion Segmentation with Tiramisu and 2.5D Stacked Slices
TLDR
A fully convolutional densely connected network (Tiramisu) for multiple sclerosis (MS) lesion segmentation using stacked slices from all three anatomical planes to achieve a 2.5D method that can make more accurate segmentation by combining information of different forms.
Assessment of Artificial Intelligence Automatic Multiple Sclerosis Lesion Delineation Tool for Clinical Use
TLDR
A white matter hyperintensity segmentation model, based on a 2D convolutional neural network, using the conventional T2-weighted fluid attenuated inversion recovery (FLAIR) MRI sequence as input was implemented, able to delineate MS lesions with a high clinical value as rated by delineation experts while outperforming popular out of the box applications.
Towards broader generalization of deep learning methods for multiple sclerosis lesion segmentation
TLDR
DeepLesionBrain improves the segmentation performance and generalization over classical techniques, and thus proposes a robust approach better suited for clinical practice.
A multi-sequences MRI deep framework study applied to glioma classfication
TLDR
This research focuses on the performance of different pre-trained deep learning models with respect to different MRI sequences, and highlights the best combinations of such model-MRI sequence couple for the specific task of classifying healthy brain against brain with glioma.
Multiple Sclerosis Lesion Segmentation - A Survey of Supervised CNN-Based Methods
TLDR
This survey investigates the supervised CNN-based methods for MS lesion segmentation and decouple these reviewed works into their algorithmic components and reports comparisons between their results.
An Empirical Study of Deep Neural Networks for Glioma Detection from MRI Sequences
TLDR
This work proposes a global framework using convolutional neural networks to create an intelligent assistant system for neurologists to diagnose the brain gliomas, and studies the performance of different neural networks on four MRI modalities.
...
...

References

SHOWING 1-10 OF 12 REFERENCES
Multiple Sclerosis Lesion Segmentation from Brain MRI via Fully Convolutional Neural Networks
TLDR
A fully convolutional neural network (CNN) based method to segment white matter lesions from multi-contrast MR images for multiple sclerosis and significant improvement in segmentation quality over the competing methods is demonstrated.
Automated Segmentation of Multiple Sclerosis Lesions Using Multi-dimensional Gated Recurrent Units
TLDR
It is demonstrated that using data augmentation, selective sampling, residual learning and/or DropConnect on the RNN state can produce better segmentation results, and it is shown that a setup which combines these techniques can outperform the state of the art in automated lesion segmentation.
Longitudinal multiple sclerosis lesion segmentation: Resource and challenge
MRI in multiple sclerosis: current status and future prospects
Asymmetric Similarity Loss Function to Balance Precision and Recall in Highly Unbalanced Deep Medical Image Segmentation
TLDR
A patch-wise 3D densely connected network with an asymmetric loss function, where large overlapping image patches for intrinsic and extrinsic data augmentation, a patch selection algorithm, and a patch prediction fusion strategy based on B-spline weighted soft voting to take into account the uncertainty of prediction in patch borders are developed.
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
TLDR
The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts and performs on-the-fly elastic deformations for efficient data augmentation during training.
Dropout: a simple way to prevent neural networks from overfitting
TLDR
It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
DEMYELINATING diseases.
G
  • ‹$ = = > € } $ ? # # $ $ @= K { $ Learning Research
  • 2014
...
...