Learning Based Registration

More often than not, medical applications require that the employed images are on the same coordinate system, making the registration process a necessary prerequisite. On this research axis we are focusing on how to utilize learning based registration methods on medical data across diferent patients as well as across time. We proposed a number of methods for registration of both CT scans as well as MRI scans accross different tissue times. These methods have also been extended to sattelite images and have been used to quantify changes.

Related Publications

Exploring Deep Registration Latent Spaces
Exploring Deep Registration Latent Spaces

MICCAI 2021
Deep learning-based concurrent brain registration and tumor segmentation
Deep learning-based concurrent brain registration and tumor segmentation

Frontiers in Computational Neuroscience 2020
Image registration of satellite imagery with deep convolutional neural networks
Image registration of satellite imagery with deep convolutional neural networks

IGARSS 2019
Linear and Deformable Image Registration with 3D Convolutional Neural Networks
Linear and Deformable Image Registration with 3D Convolutional Neural Networks

MICCAI RAMBO Workshop 2018