Medical Image Registration Using Deep Neural Networks: A Comprehensive Review

  title={Medical Image Registration Using Deep Neural Networks: A Comprehensive Review},
  author={Hamid Reza Boveiri and Raouf Khayami and Reza Javidan and Ali Reza Mehdizadeh},
  journal={Comput. Electr. Eng.},

A deep learning based framework for the registration of three dimensional multi-modal medical images of the head

A registration framework is introduced that creates synthetic data to augment existing datasets, generates ground truth data to be used in the training and testing of algorithms, and registers multi-modal images in an accurate and fast manner and automatically classifies the image modality so that the process of registration can be fully automated.

Is image-to-image translation the panacea for multimodal image registration? A comparative study

The results suggest that, although I2I translation may be helpful when the modalities to register are clearly correlated, registration of modalities which express distinctly different properties of the sample are not well handled by the I 2I translation approach.

Atlas-Based Segmentation of the Human Embryo Using Deep Learning with Minimal Supervision

We propose an atlas-based segmentation framework to achieve segmentation and spatial alignment of three-dimensional ultrasound images of the human embryo acquired during the first trimester of

An Infrared-Visible Image Registration Method Based on the Constrained Point Feature

Registration results demonstrate that the proposed framework outperforms five state-of-the-art registration algorithms in terms of accuracy, speed, and robustness.

An Improved Fourier-Mellin Transform-Based Registration Used in TDI-CMOS

An improved Fourier-Mellin transform-based registration method is presented, which can be used to realize the registration-based time-delayed integration (TDI), and the TDI images generated by the proposed framework have higher quality in the index of information entropy, average gradient, and spatial frequency response.

Unsupervised Deformable Image Registration Using Polyphase UNet for 3D Brain MRI Volumes

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Medical image registration using unsupervised deep neural network: A scoping literature review

A narrative review on radiotherapy practice in the era of artificial intelligence: how relevant is the medical physicist?

To achieve successful implementation of AI in radiotherapy and optimize radiotherapy procedures, clinical medical physicist should receive some compulsory training in AI technologies during their education and training.

Multi-Atlas Segmentation and Spatial Alignment of the Human Embryo in First Trimester 3D Ultrasound

A multi-atlas framework for automatic segmentation and spatial alignment of the embryo using deep learning with minimal supervision can accurately segment and spatially align the embryo in 3D ultrasound images and is robust to the variation in quality that existed in the available atlases.



A survey on deep learning in medical image analysis

Learning Rigid Image Registration - Utilizing Convolutional Neural Networks for Medical Image Registration

This paper describes and evaluates the use of convolutional neural networks for both mono- and multi- modality registration and compares their performance to more traditional schemes, namely multi-scale, iterative registration and incorporates inverse consistency of the learned spatial transformations to impose additional constraints on the network during training.

An unsupervised network for fast microscopic image registration

An unsupervised network for image registration is proposed and image deformation is achieved by resampling, which can make deformation step derivable and the network optimizes its parameters directly by minimizing the loss between registered image and reference image without ground truth.

Adversarial Image Registration with Application for MR and TRUS Image Fusion

By training two deep neural networks simultaneously, one being a generator and the other being a discriminator, one can obtain not only a network for image registration, but also a metric network which can help evaluate the quality of image registration.

Deep similarity learning for multimodal medical images

A novel deep similarity learning method that trains a binary classifier to learn the correspondence of two image patches to show the advantage of the proposed metric for a highly accurate and robust similarity measure.

Image synthesis-based multi-modal image registration framework by using deep fully convolutional networks

An image synthesis-based multi-modal image registration framework that can capture the complex nonlinear relationship between different modalities and discover complex structural representations automatically by a large number of trainable mapping and parameters and perform accurate image synthesis is proposed.

On the Adaptability of Unsupervised CNN-Based Deformable Image Registration to Unseen Image Domains

This work explores the adaptability of CNN-based image registration to different organs/modalities, and considers a simple transfer learning strategy to study the generalisation of such model to unseen target domains, and devise a one-shot learning scheme taking advantage of the unsupervised nature of the proposed method.