3-D Reconstruction in Canonical Co-Ordinate Space From Arbitrarily Oriented 2-D Images

  title={3-D Reconstruction in Canonical Co-Ordinate Space From Arbitrarily Oriented 2-D Images},
  author={Benjamin Hou and B. Khanal and Amir Alansary and Steven G. McDonagh and Alice Davidson and Mary A. Rutherford and Joseph V. Hajnal and Daniel Rueckert and Ben Glocker and Bernhard Kainz},
  journal={IEEE Transactions on Medical Imaging},
Limited capture range, and the requirement to provide high quality initialization for optimization-based 2-D/3-D image registration methods, can significantly degrade the performance of 3-D image reconstruction and motion compensation pipelines. Challenging clinical imaging scenarios, which contain significant subject motion, such as fetal in-utero imaging, complicate the 3-D image and volume reconstruction process. In this paper, we present a learning-based image registration method capable of… 

Real-Time Deep Pose Estimation With Geodesic Loss for Image-to-Template Rigid Registration

The proposed deep learning-based methods that are trained to find the 3-D position of arbitrarily-oriented subjects or anatomy in a canonical space based on slices or volumes of medical images can dramatically enhance the performance of automatic imaging devices and image processing methods of the future.

Real-time Deep Registration With Geodesic Loss

The results show that in such registration applications that are amendable to learning, the proposed deep learning methods with geodesic loss minimization can achieve accurate results with a wide capture range in real-time, and a combination of the proposed methods with optimization-based registration algorithms can dramatically enhance the performance of automatic imaging devices and image processing methods of the future.

Transformed Grid Distance Loss for Supervised Image Registration

The proposed Transformed Grid Distance loss solves the problem of rotation representation but unites the gap between translation and rotation and demonstrates superior performance in comparison with conventional losses while requiring no alteration to the network input, output, or network structure at all.


Qualitative and quantitative evaluations confirm that the proposed learning-based rigid registration method can achieve accurate 2DUS-3DMRI rigid registration.

SVoRT: Iterative Transformer for Slice-to-Volume Registration in Fetal Brain MRI

This work proposes a novel slice-to-volume registration method using Transformers trained on syn-thetically transformed data, which model multiple stacks of MR slices as a sequence and achieves lower registration error and better reconstruction quality compared with existing state-of-the-art methods.

Deep Cascade Networks for Single 2D US Slice to 3D CT/MRI Image Registration

The results show the efficiency of the proposed registration networks, which have the potential to improve the robustness and accuracy of intraoperative patient registration.

Adaptive 3D Localization of 2D Freehand Ultrasound Brain Images

AdLocUI is a framework thatizes 2D U ltrasound I mages in the 3D anatomical atlas without using any external tracking sensor, and achieves significantly better localization accuracy than the baselines.

Volumetric MRI reconstruction from 2D slices in the presence of motion

Methods to volumetrically reconstruct geometrically consistent, high-resolution (HR) three-dimensional (3D) images from motion-corrupted, possibly sparse, low-resolution 2D MR slices are presented.

Deep Pose Estimation for Image-Based Registration

A general Riemannian formulation of the pose estimation problem is proposed and it is shown that it can improve registration accuracy for image-based 2D to 3D registration and compare its performance with popular and most commonly used existing methods.



Predicting Slice-to-Volume Transformation in Presence of Arbitrary Subject Motion

A regression approach that learns to predict rotations and translations of arbitrary 2D image slices from 3D volumes, with respect to a learned canonical atlas co-ordinate system, which is a general solution to the 2D/3D initialization problem.

A CNN Regression Approach for Real-Time 2D/3D Registration

The proposed CNN regression approach has been quantitatively evaluated on 3 potential clinical applications, demonstrating its significant advantage in providing highly accurate real-time 2-D/3-D registration with a significantly enlarged capture range when compared to intensity-based methods.

Intersection Based Motion Correction of Multislice MRI for 3-D in Utero Fetal Brain Image Formation

An alternative approach to slice intersection motion correction (SIMC), that seeks to directly co-align multiple slice stacks by considering the matching structure along all intersecting slice pairs in all orthogonally planned slices that are acquired in clinical imaging studies, is described.

Non-rigid Craniofacial 2D-3D Registration Using CNN-Based Regression

A CNN regression based non-rigid 2D-3D registration method is proposed to update the reference volumetric image and the digitally-reconstructed-radiograph (DRR) for convergence to the target X-ray image.

A Unified Approach to Diffusion Direction Sensitive Slice Registration and 3-D DTI Reconstruction From Moving Fetal Brain Anatomy

This paper presents an approach to 3-D diffusion tensor image (DTI) reconstruction from multi-slice diffusion weighted (DW) magnetic resonance imaging acquisitions of the moving fetal brain, showing improved rendition of cortical anatomy and extraction of white matter tracts.

Slice-to-volume medical image registration: a survey

PVR: Patch-to-Volume Reconstruction for Large Area Motion Correction of Fetal MRI

A novel method for the correction of motion artifacts that are present in fetal magnetic resonance imaging (MRI) scans of the whole uterus is presented and an efficient parallel implementation of PVR is described allowing its execution within reasonable time on commercially available graphics processing units, enabling its use in the clinical practice.

PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization

This work trains a convolutional neural network to regress the 6-DOF camera pose from a single RGB image in an end-to-end manner with no need of additional engineering or graph optimisation, demonstrating that convnets can be used to solve complicated out of image plane regression problems.

MRI of Moving Subjects Using Multislice Snapshot Images With Volume Reconstruction (SVR): Application to Fetal, Neonatal, and Adult Brain Studies

A novel methodology, Snapshot magnetic resonance imaging with volume reconstruction (SVR) has been developed for imaging moving subjects at high resolution and high signal-to-noise ratio (SNR) and fine structure of the in-utero fetal brain is clearly revealed for the first time.