Intensity non-uniformity correction in MR imaging using residual cycle generative adversarial network

Abstract

Correcting or reducing the effects of voxel intensity non-uniformity (INU) within a given tissue type is a crucial issue for quantitative magnetic resonance (MR) image analysis in daily clinical practice. Although having no severe impact on visual diagnosis, the INU can highly degrade the performance of automatic quantitative analysis such as segmentation, registration, feature extraction and radiomics. In this study, we present an advanced deep learning based INU correction algorithm called residual cycle generative adversarial network (res-cycle GAN), which integrates the residual block concept into a cycle-consistent GAN (cycle-GAN). In cycle-GAN, an inverse transformation was implemented between the INU uncorrected and corrected magnetic resonance imaging (MRI) images to constrain the model through forcing the calculation of both an INU corrected MRI and a synthetic corrected MRI. A fully convolution neural network integrating residual blocks was applied in the generator of cycle-GAN to enhance end-to-end raw MRI to INU corrected MRI transformation. A cohort of 55 abdominal patients with T1-weighted MR INU images and their corrections with a clinically established and commonly used method, namely, N4ITK were used as a pair to evaluate the proposed res-cycle GAN based INU correction algorithm. Quantitatively comparisons of normalized mean absolute error (NMAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC) indices, and spatial non-uniformity (SNU) were made among the proposed method and other approaches. Our res-cycle GAN based method achieved an NMAE of 0.011 ± 0.002, a PSNR of 28.0 ± 1.9 dB, an NCC of 0.970 ± 0.017, and a SNU of 0.298 ± 0.085. Our proposed method has significant improvements (p < 0.05) in NMAE, PSNR, NCC and SNU over other algorithms including conventional GAN and U-net. Once the model is well trained, our approach can automatically generate the corrected MR images in a few minutes, eliminating the need for manual setting of parameters.

A review of deep learning based methods for medical image multi-organ segmentation

“…Inspired by the success of DL in computer vision, researchers have proposed various methods to extend the use of DL techniques to medical imaging. To date, DL has been extensively studied in medical image segmentation , image synthesis , image enhancement and correction [97][98][99][100][101][102][103][104][105][106][107], and registration [108][109][110][111][112][113][114][115][116][117][118][119][120][121]. DL-based multi-organ segmentation technique represents a significant potential in daily practices of radiation therapy since it can expedite the contouring process, improve contour accuracy and consistency and promote compliance to delineation guidelines [39,45,[122][123][124][125][126].…”

A transfer learning approach to facilitate ComBat-based harmonization of multicentre radiomic features in new datasets

“…However, it has been shown recently that although such an approach can help towards reducing multicentre effects, it may still be insufficient to fully compensate them [43,45]. Techniques based on deep learning (convolutional neural networks, CNN or generative adversarial networks, GAN and their variants) have also been considered in order to standardize or harmonize medical images [44,[46][47][48], including with an evaluation of the impact on resulting radiomic features, in the context of lung lesions in CT images [46]. Another paper [49] showed in a proposed workflow evaluating harmonization techniques using synthetic and real data comparing ComBat and cycleGaN that both methods perform well for removing various types of noises while preserving manually added synthesis lesions, but also for removing site effects on data coming from 2…”

“…ComBat recently outperformed 6 other methods for batch effect removal using microarray datasets from brain RNA samples and two simulated datasets [25]. Although an extensive comparison of ComBat with other methods remains to be carried out explicitly in the context of radiomics, it has already been identified as a promising technique and is being increasingly and successfully used in recent radiomics studies [15,20,22,24,32,[51][52][53][54][55][56]. It however has some limitations regarding its use in practice and we previously addressed two of these with the proposed BM-ComBat to allow for more flexibility in choosing a reference label and improving the estimation [44].…”

Self‐supervised learning for accelerated 3D high‐resolution ultrasound imaging

“…Inspired by the tremendous success of deep learning in computer vision, 24–27 deep learning‐based methods have been recently investigated for medical image reconstruction, 28–31 analysis, 32–36 and synthesis 37,38 . Studies have demonstrated deep learning‐based approaches significantly outperform over CS‐based methods for image reconstruction 39‐42 .…”

“…18,19 Beamforming techniques have been developed to reconstruct 3D US images from a limited number of measurements, where a single focused line is virtually modeled using multiple adjacent transducer elements. [20][21][22][23] Inspired by the tremendous success of deep learning in computer vision, [24][25][26][27] deep learning-based methods have been recently investigated for medical image reconstruction, [28][29][30][31] analysis, [32][33][34][35][36] and synthesis. 37,38 Studies have demonstrated deep learning-based approaches significantly outperform over CS-based methods for image reconstruction.…”