Higher SNR PET image prediction using a deep learning model and MRI image

Liu, Chih-Chieh; Qi, Jinyi

doi:10.1088/1361-6560/ab0dc0

Cited by 56 publications

(35 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are several novel aspects of this study. The majority of other related work has been performed in 2D since processing times are faster, fewer GPU memory issues are encountered, and the availability of pretrained 2D networks provides additional options in the choice of training objectives [ 23 – 26 ]. However, volumetric 3D PET data are the medical standard, and inclusion of the additional dimension of data was expected to improve training stability and robustness of the network performance [ 27 ].…”

Section: Discussionmentioning

confidence: 99%

“…The noise in PET images is generally assumed to follow Gaussian and/or Poisson distributions, and deep learning is especially well positioned to address this since the characteristic features of the noise, regardless of the assumed model, are inherently learned through training. Several techniques have previously been applied successfully for denoising PET images [ 23 – 26 ], to date however, there are very few studies exploring CNNs which handle 3D data. Due to the volumetric nature of data, it is expected that the performance of CNNs could be improved [ 27 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Convolutional neural networks for improving image quality with noisy PET data

et al. 2020

View full text Add to dashboard Cite

Goal PET is a relatively noisy process compared to other imaging modalities, and sparsity of acquisition data leads to noise in the images. Recent work has focused on machine learning techniques to improve PET images, and this study investigates a deep learning approach to improve the quality of reconstructed image volumes through denoising by a 3D convolution neural network. Potential improvements were evaluated within a clinical context by physician performance in a reading task. Methods A wide range of controlled noise levels was emulated from a set of chest PET data in patients with lung cancer, and a convolutional neural network was trained to denoise the reconstructed images using the full-count reconstructions as the ground truth. The benefits, over conventional Gaussian smoothing, were quantified across all noise levels by observer performance in an image ranking and lesion detection task. Results The CNN-denoised images were generally ranked by the physicians equal to or better than the Gaussian-smoothed images for all count levels, with the largest effects observed in the lowest-count image sets. For the CNN-denoised images, overall lesion contrast recovery was 60% and 90% at the 1 and 20 million count levels, respectively. Notwithstanding the reduced lesion contrast recovery in noisy data, the CNN-denoised images also yielded better lesion detectability in low count levels. For example, at 1 million true counts, the average true positive detection rate was around 40% for the CNN-denoised images and 30% for the smoothed images. Conclusion Significant improvements were found for CNN-denoising for very noisy images, and to some degree for all noise levels. The technique presented here offered however limited benefit for detection performance for images at the count levels routinely encountered in the clinic.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Convolutional neural networks for improving image quality with noisy PET data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…A three‐dimensional‐simulated [ 18 F]fluorodeoxyglucose (FDG) PET phantom was constructed with a voxel‐side length of 1 mm, based on the BrainWeb segmented MR database. The gray and white matter tissue classes were assigned intensities with the ratio of 4:1 in keeping with the expected uptake from an FDG tracer . Real PET images have more structural variation than the produced piecewise constant simulated phantom.…”

Section: Methodsmentioning

confidence: 99%

Intercomparison of MR‐informed PET image reconstruction methods

et al. 2019

View full text Add to dashboard Cite

Purpose Numerous image reconstruction methodologies for positron emission tomography (PET) have been developed that incorporate magnetic resonance (MR) imaging structural information, producing reconstructed images with improved suppression of noise and reduced partial volume effects. However, the influence of MR structural information also increases the possibility of suppression or bias of structures present only in the PET data (PET‐unique regions). To address this, further developments for MR‐informed methods have been proposed, for example, through inclusion of the current reconstructed PET image, alongside the MR image, in the iterative reconstruction process. In this present work, a number of kernel and maximum a posteriori (MAP) methodologies are compared, with the aim of identifying methods that enable a favorable trade‐off between the suppression of noise and the retention of unique features present in the PET data. Methods The reconstruction methods investigated were: the MR‐informed conventional and spatially compact kernel methods, referred to as KEM and KEM largest value sparsification (LVS) respectively; the MR‐informed Bowsher and Gaussian MR‐guided MAP methods; and the PET‐MR‐informed hybrid kernel and anato‐functional MAP methods. The trade‐off between improving the reconstruction of the whole brain region and the PET‐unique regions was investigated for all methods in comparison with postsmoothed maximum likelihood expectation maximization (MLEM), evaluated in terms of structural similarity index (SSIM), normalized root mean square error (NRMSE), bias, and standard deviation. Both simulated BrainWeb (10 noise realizations) and real [18F] fluorodeoxyglucose (FDG) three‐dimensional datasets were used. The real [18F]FDG dataset was augmented with simulated tumors to allow comparison of the reconstruction methodologies for the case of known regions of PET‐MR discrepancy and evaluated at full counts (100%) and at a reduced (10%) count level. Results For the high‐count simulated and real data studies, the anato‐functional MAP method performed better than the other methods under investigation (MR‐informed, PET‐MR‐informed and postsmoothed MLEM), in terms of achieving the best trade‐off for the reconstruction of the whole brain and PET‐unique regions, assessed in terms of the SSIM, NRMSE, and bias vs standard deviation. The inclusion of PET information in the anato‐functional MAP method enables the reconstruction of PET‐unique regions to attain similarly low levels of bias as unsmoothed MLEM, while moderately improving the whole brain image quality for low levels of regularization. However, for low count simulated datasets the anato‐functional MAP method performs poorly, due to the inclusion of noisy PET information in the regularization term. For the low counts simulated dataset, KEM LVS and to a lesser extent, HKEM performed better than the other methods under investigation in terms of achieving the best trade‐off for the reconstruction of the whole brain and PET‐unique regions, assessed in terms of t...

show abstract

“…The gray and white matter tissue classes were assigned intensities with the ratio of 4:1 in keeping with the expected uptake from an FDG tracer. 30,60,61 Real PET images have more structural variation than the produced piecewise constant simulated phantom. To discourage overly piecewise constant images, Gaussian smoothed random structures were incorporated into the simulated PET phantom, in accordance with Eq.…”

Section: B Simulation Studiesmentioning

confidence: 99%

Intercomparison of MR-Informed Methods for PET Image Reconstruction

Bland

Mehranian

Belzunce

et al. 2018

2018 IEEE Nuclear Science Symposium and Medical Imaging Conference Proceedings (NSS/MIC)

View full text Add to dashboard Cite

Purpose: Numerous image reconstruction methodologies for positron emission tomography (PET) have been developed that incorporate magnetic resonance (MR) imaging structural information, producing reconstructed images with improved suppression of noise and reduced partial volume effects. However, the influence of MR structural information also increases the possibility of suppression or bias of structures present only in the PET data (PET-unique regions). To address this, further developments for MR-informed methods have been proposed, for example, through inclusion of the current reconstructed PET image, alongside the MR image, in the iterative reconstruction process. In this present work, a number of kernel and maximum a posteriori (MAP) methodologies are compared, with the aim of identifying methods that enable a favorable trade-off between the suppression of noise and the retention of unique features present in the PET data. Methods: The reconstruction methods investigated were: the MR-informed conventional and spatially compact kernel methods, referred to as KEM and KEM largest value sparsification (LVS) respectively; the MR-informed Bowsher and Gaussian MR-guided MAP methods; and the PET-MR-informed hybrid kernel and anato-functional MAP methods. The trade-off between improving the reconstruction of the whole brain region and the PET-unique regions was investigated for all methods in comparison with postsmoothed maximum likelihood expectation maximization (MLEM), evaluated in terms of structural similarity index (SSIM), normalized root mean square error (NRMSE), bias, and standard deviation. Both simulated BrainWeb (10 noise realizations) and real [ 18 F] fluorodeoxyglucose (FDG) threedimensional datasets were used. The real [ 18 F]FDG dataset was augmented with simulated tumors to allow comparison of the reconstruction methodologies for the case of known regions of PET-MR discrepancy and evaluated at full counts (100%) and at a reduced (10%) count level. Results: For the high-count simulated and real data studies, the anato-functional MAP method performed better than the other methods under investigation (MR-informed, PET-MR-informed and postsmoothed MLEM), in terms of achieving the best trade-off for the reconstruction of the whole brain and PET-unique regions, assessed in terms of the SSIM, NRMSE, and bias vs standard deviation. The inclusion of PET information in the anato-functional MAP method enables the reconstruction of PET-unique regions to attain similarly low levels of bias as unsmoothed MLEM, while moderately improving the whole brain image quality for low levels of regularization. However, for low count simulated datasets the anato-functional MAP method performs poorly, due to the inclusion of noisy PET information in the regularization term. For the low counts simulated dataset, KEM LVS and to a lesser extent, HKEM performed better than the other methods under investigation in terms of achieving the best trade-off for the reconstruction of the whole brain and PET-unique regions, assessed in te...

show abstract

Higher SNR PET image prediction using a deep learning model and MRI image

Cited by 56 publications

References 47 publications

Convolutional neural networks for improving image quality with noisy PET data

Convolutional neural networks for improving image quality with noisy PET data

Intercomparison of MR‐informed PET image reconstruction methods

Intercomparison of MR-Informed Methods for PET Image Reconstruction

Contact Info

Product

Resources

About