Dynamic PET Image Denoising Using Deep Convolutional Neural Networks Without Prior Training Datasets

Hashimoto, Fumio; Ohba, Hiroyuki; Ote, Kibo; Teramoto, Akira; Tsukada, Hideo

doi:10.1109/access.2019.2929230

Cited by 107 publications

(79 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consistent with previous studies [25,26,33], the findings presented here suggest that image quality, in terms of interpixel spatial variance and measurement reproducibility, were significantly improved. However, we found that these findings might not always translate directly into clinical improvements-the uptake in small foci was sometimes not accurately quantified, and lesion detection and localization realized less benefit.…”

Section: Discussionsupporting

confidence: 92%

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: Discussionsupporting

confidence: 92%

Convolutional neural networks for improving image quality with noisy PET data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Nowadays deep learning is the most active research area, and much work has been reported in PET denoising [13][14][15][16][17][18][19]. Deep learning denoising is effective and has better results than traditional methods.…”

Section: Discussionmentioning

confidence: 99%

Poisson-noise weighted filter for time-of-flight positron emission tomography

Zeng

Huang

2020

Vis. Comput. Ind. Biomed. Art

View full text Add to dashboard Cite

Image reconstruction for list-mode time-of-flight (TOF) positron emission tomography (PET) can be achieved by analytic algorithms. The backprojection filtering (BPF) algorithm is an efficient algorithm for this task. The conventional noise control method for analytic image reconstruction is the use of a stationary lowpass filter, which does not model the Poisson noise properly. This study proposes a nonstationary filter for Poisson noise control. The filter is implemented in the spatial domain in a form similar to convolution.

show abstract

“…In DIP, no pre-training dataset is needed, a convolutional neural network (CNN) is initialized with random parameters, and only random noise is prepared as the network input. Research related to DIP has focused on natural image denoising, inpainting, super-resolution reconstruction [33,34], PET image reconstruction [35,36], and even compressed sensing recovery problems [37].…”

Section: Introductionmentioning

confidence: 99%

Reference-Driven Compressed Sensing MR Image Reconstruction Using Deep Convolutional Neural Networks without Pre-Training

Zhao

Zhang

Gan

2020

Sensors

View full text Add to dashboard Cite

Deep learning has proven itself to be able to reduce the scanning time of Magnetic Resonance Imaging (MRI) and to improve the image reconstruction quality since it was introduced into Compressed Sensing MRI (CS-MRI). However, the requirement of using large, high-quality, and patient-based datasets for network training procedures is always a challenge in clinical applications. In this paper, we propose a novel deep learning based compressed sensing MR image reconstruction method that does not require any pre-training procedure or training dataset, thereby largely reducing clinician dependence on patient-based datasets. The proposed method is based on the Deep Image Prior (DIP) framework and uses a high-resolution reference MR image as the input of the convolutional neural network in order to induce the structural prior in the learning procedure. This reference-driven strategy improves the efficiency and effect of network learning. We then add the k-space data correction step to enforce the consistency of the k-space data with the measurements, which further improve the image reconstruction accuracy. Experiments on in vivo MR datasets showed that the proposed method can achieve more accurate reconstruction results from undersampled k-space data.

show abstract

Dynamic PET Image Denoising Using Deep Convolutional Neural Networks Without Prior Training Datasets

Cited by 107 publications

References 39 publications

Convolutional neural networks for improving image quality with noisy PET data

Convolutional neural networks for improving image quality with noisy PET data

Poisson-noise weighted filter for time-of-flight positron emission tomography

Reference-Driven Compressed Sensing MR Image Reconstruction Using Deep Convolutional Neural Networks without Pre-Training

Contact Info

Product

Resources

About