3D Deformable Convolutions for MRI classification

Marina Pominova1Ekaterina Kondrateva1Maxim Sharaev1Sergey Pavlov1Alexander Bernstein1Evgeny Burnaev1

1Skolkovo Institute of Science and Technology2Moscow Institute of Physics and Technology

18th IEEE International Conference On Machine Learning And Applications 2019

Abstract

Deep learning convolution neural networks have proved to be a powerful tool for MRI analysis. In current work, we explore the potential of the deformable convolution deep neural network layers for MRI data classification. We propose new 3D deformable convolutions (d-convolutions), implement them in VoxResNet architecture and apply for structural MRI data classification. We show that 3D d-convolutions outperform standard ones and are effective for unprocessed 3D MR images being robust to particular geometrical properties of the data. Firstly proposed dVoxResNet architecture exhibits high potential for the use in MRI data classification.

Materials

Paper

Code

Copy bibtex

Contact

If you have any questions about this work, please contact us under ekaterina.kondrateva@skoltech.ru.