Detection and Correction of Cardiac MRI Motion Artefacts During Reconstruction from k-space

Abstract

Segmenting anatomical structures in medical images has been successfully addressed with deep learning methods for a range of applications. However, this success is heavily dependent on the quality of the image that is being segmented. A commonly neglected point in the medical image analysis community is the vast amount of clinical images that have severe image artefacts due to organ motion, movement of the patient and/or image acquisition related issues. In this paper, we discuss the implications of image motion artefacts on cardiac MR segmentation and compare a variety of approaches for jointly correcting for artefacts and segmenting the cardiac cavity. The method is based on our recently developed joint artefact detection and reconstruction method, which reconstructs high quality MR images from k-space using a joint loss function and essentially converts the artefact correction task to an under-sampled image reconstruction task by enforcing a data consistency term. In this paper, we propose to use a segmentation network coupled with this in an end-to-end framework. Our training optimises three different tasks: 1) image artefact detection, 2) artefact correction and 3) image segmentation. We train the reconstruction network to automatically correct for motion-related artefacts using synthetically corrupted cardiac MR k-space data and uncorrected reconstructed images. Using a test set of 500 2D+time cine MR acquisitions from the UK Biobank data set, we achieve demonstrably good image quality and high segmentation accuracy in the presence of synthetic motion artefacts. We quantitatively compare our method with a variety of techniques for jointly recovering image quality and performing image segmentation. We showcase better performance compared to state-of-the-art image correction techniques. Moreover, our method preserves the quality of uncorrupted images and therefore can be utilised as a global image reconstruction algorithm.

A domain-agnostic MR reconstruction framework using a randomly weighted neural network

“…Similarly, several methods such as [28,29,30,31] optimize the weights of a Convolutional Neural Network (CNN) for MR image reconstruction. Recently developed methods such as [32,33,34] optimize a Generative Adversarial Network (GAN) for MRI reconstruction. Different from these, a local-global recurrent neural network [35] uses a bidirectional LSTM that replaces the dense network structure of AUTOMAP [27] for removing aliasing artifacts in the reconstructed image.…”

Image-based Motion Artifact Reduction on Liver Dynamic Contrast Enhanced MRI

“…Recently, deep learning (DL) approaches have been developed in medical imaging use cases including image reconstruction and artifact reduction (11,12), motion detection and correction (13), and image quality control (14,15). DL methods utilized convolutional neural networks (CNNs) to extract features of different types of artifacts and correct them in brain (16,17), abdominal (18)(19)(20) and cardiac imaging (13).…”

“…Recently, deep learning (DL) approaches have been developed in medical imaging use cases including image reconstruction and artifact reduction (11,12), motion detection and correction (13), and image quality control (14,15). DL methods utilized convolutional neural networks (CNNs) to extract features of different types of artifacts and correct them in brain (16,17), abdominal (18)(19)(20) and cardiac imaging (13). Specifically for liver DCE-MRI, Tamada et al proposed a denoising CNN on multi-phase magnitude-only image patches that learned the artifact patterns as residual feature maps and then subtracted them from the original images to obtain the motion reduced images (19).…”

Dilated convolution network with edge fusion block and directional feature maps for cardiac MRI segmentation

“…In recent years, with the development of deep convolutional networks (CNNs), many natural image segmentation (Cheng and Li, 2021;Aganj and Fischl, 2021) and medical image segmentation (Pang et al 2021;Oksuz et al 2020) methods have been proposed in the field of computer vision and achieved great success. U-Net Ronneberger et al (2015) is one of the seminal works in medical image segmentation task.…”

“…The encoder can automatically bend an inaccurate heart shape to a close but correct shape. Oksuz et al (2020) propose a network that could automatically correct motion-related artifacts, and the network achieved good image quality and high segmentation accuracy in the presence of synthetic motion. Yang et al (2021) propose a deep dilated block adversarial network, which uses the properties of dilated convolution to acquire and connect multiscale features.…”