Dose reduction and image enhancement in micro-CT using deep learning.
convolutional neural networks
deep learning
dose reduction
image denoising
micro-computed tomography
Journal
Medical physics
ISSN: 2473-4209
Titre abrégé: Med Phys
Pays: United States
ID NLM: 0425746
Informations de publication
Date de publication:
Sep 2023
Sep 2023
Historique:
revised:
14
02
2023
received:
08
09
2022
accepted:
09
03
2023
medline:
11
9
2023
pubmed:
31
3
2023
entrez:
30
3
2023
Statut:
ppublish
Résumé
In preclinical settings, micro-computed tomography (CT) provides a powerful tool to acquire high resolution anatomical images of rodents and offers the advantage to in vivo non-invasively assess disease progression and therapy efficacy. Much higher resolutions are needed to achieve scale-equivalent discriminatory capabilities in rodents as those in humans. High resolution imaging however comes at the expense of increased scan times and higher doses. Specifically, with preclinical longitudinal imaging, there are concerns that dose accumulation may affect experimental outcomes of animal models. Dose reduction efforts under the ALARA (as low as reasonably achievable) principles are thus a key point of attention. However, low dose CT acquisitions inherently induce higher noise levels which deteriorate image quality and negatively impact diagnostic performance. Many denoising techniques already exist, and deep learning (DL) has become increasingly popular for image denoising, but research has mostly focused on clinical CT with limited studies conducted on preclinical CT imaging. We investigate the potential of convolutional neural networks (CNN) for restoring high quality micro-CT images from low dose (noisy) images. The novelty of the CNN denoising frameworks presented in this work consists of utilizing image pairs with realistic CT noise present in the input as well as the target image used for the model training; a noisier image acquired with a low dose protocol is matched to a less noisy image acquired with a higher dose scan of the same mouse. Low and high dose ex vivo micro-CT scans of 38 mice were acquired. Two CNN models, based on a 2D and 3D four-layer U-Net, were trained with mean absolute error (30 training, 4 validation and 4 test sets). To assess denoising performance, ex vivo mice and phantom data were used. Both CNN approaches were compared to existing methods, like spatial filtering (Gaussian, Median, Wiener) and iterative total variation image reconstruction algorithm. Image quality metrics were derived from the phantom images. A first observer study (n = 23) was set-up to rank overall quality of differently denoised images. A second observer study (n = 18) estimated the dose reduction factor of the investigated 2D CNN method. Visual and quantitative results show that both CNN algorithms exhibit superior performance in terms of noise suppression, structural preservation and contrast enhancement over comparator methods. The quality scoring by 23 medical imaging experts also indicates that the investigated 2D CNN approach is consistently evaluated as the best performing denoising method. Results from the second observer study and quantitative measurements suggest that CNN-based denoising could offer a 2-4× dose reduction, with an estimated dose reduction factor of about 3.2 for the considered 2D network. Our results demonstrate the potential of DL in micro-CT for higher quality imaging at low dose acquisition settings. In the context of preclinical research, this offers promising future prospects for managing the cumulative severity effects of radiation in longitudinal studies.
Sections du résumé
BACKGROUND
BACKGROUND
In preclinical settings, micro-computed tomography (CT) provides a powerful tool to acquire high resolution anatomical images of rodents and offers the advantage to in vivo non-invasively assess disease progression and therapy efficacy. Much higher resolutions are needed to achieve scale-equivalent discriminatory capabilities in rodents as those in humans. High resolution imaging however comes at the expense of increased scan times and higher doses. Specifically, with preclinical longitudinal imaging, there are concerns that dose accumulation may affect experimental outcomes of animal models.
PURPOSE
OBJECTIVE
Dose reduction efforts under the ALARA (as low as reasonably achievable) principles are thus a key point of attention. However, low dose CT acquisitions inherently induce higher noise levels which deteriorate image quality and negatively impact diagnostic performance. Many denoising techniques already exist, and deep learning (DL) has become increasingly popular for image denoising, but research has mostly focused on clinical CT with limited studies conducted on preclinical CT imaging. We investigate the potential of convolutional neural networks (CNN) for restoring high quality micro-CT images from low dose (noisy) images. The novelty of the CNN denoising frameworks presented in this work consists of utilizing image pairs with realistic CT noise present in the input as well as the target image used for the model training; a noisier image acquired with a low dose protocol is matched to a less noisy image acquired with a higher dose scan of the same mouse.
METHODS
METHODS
Low and high dose ex vivo micro-CT scans of 38 mice were acquired. Two CNN models, based on a 2D and 3D four-layer U-Net, were trained with mean absolute error (30 training, 4 validation and 4 test sets). To assess denoising performance, ex vivo mice and phantom data were used. Both CNN approaches were compared to existing methods, like spatial filtering (Gaussian, Median, Wiener) and iterative total variation image reconstruction algorithm. Image quality metrics were derived from the phantom images. A first observer study (n = 23) was set-up to rank overall quality of differently denoised images. A second observer study (n = 18) estimated the dose reduction factor of the investigated 2D CNN method.
RESULTS
RESULTS
Visual and quantitative results show that both CNN algorithms exhibit superior performance in terms of noise suppression, structural preservation and contrast enhancement over comparator methods. The quality scoring by 23 medical imaging experts also indicates that the investigated 2D CNN approach is consistently evaluated as the best performing denoising method. Results from the second observer study and quantitative measurements suggest that CNN-based denoising could offer a 2-4× dose reduction, with an estimated dose reduction factor of about 3.2 for the considered 2D network.
CONCLUSIONS
CONCLUSIONS
Our results demonstrate the potential of DL in micro-CT for higher quality imaging at low dose acquisition settings. In the context of preclinical research, this offers promising future prospects for managing the cumulative severity effects of radiation in longitudinal studies.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
5643-5656Informations de copyright
© 2023 American Association of Physicists in Medicine.
Références
Cunha L, Horvath I, Ferreira S, et al. Preclinical imaging: an essential ally in modern biosciences. Mol Diagn Ther. 2014;18(2):153-173.
Russell WM, Burch RL. The Principles of Humane Experimental Technique. Methuen & CO LTD; 1959.
Vanhove C, Bankstahl JP, Krämer DS, Visser E, Belcari N, Vandenberghe S. Accurate molecular imaging of small animals taking into account animal models, handling, anaesthesia, quality control and imaging system performance. EJNMMI Phys. 2015;2(31).
McCollough CH, Bushberg JT, Fletcher JG, Eckel LJ. Answers to common questions about the use and safety of CT scans. Mayo Clin Proc. 2015;90(10):1380-1392.
Molinos C, Sasser T, Salmon P, et al. Low-dose imaging in a new preclinical total-body PET/CT scanner. Front Med (Lausanne). 2019;6:88.
Taschereau R, Chatziioannou AF. Monte Carlo simulations of absorbed dose in a mouse phantom from 18-fluorine compounds. Med Phys. 2007;34(3):1026-1036.
Wang J, Lu H, Li T, Liang Z. Sinogram noise reduction for low-dose CT by statistics-based nonlinear filters. In: Fitzpatrick JM, Reinhardt JM, eds. Medical Imaging 2005: Image Processing. SPIE; 2005:2058-2066.
Liu J, Ma J, Zhang Y, et al. Discriminative feature representation to improve projection data inconsistency for low dose CT imaging. IEEE Trans Med Imaging. 2017;36(12):2499-2509.
Geyer LL, Schoepf UJ, Meinel FG, et al. State of the art: iterative CT reconstruction techniques. Radiology. 2015;276(2):339-357.
Willemink MJ, Leiner T, de Jong PA, et al. Iterative reconstruction techniques for computed tomography part 2: initial results in dose reduction and image quality. Eur Radiol. 2013;23(6):1632-1642.
Diwakar M, Kumar M. A review on CT image noise and its denoising. Biomed Signal Process Control. 2018;42:73-88.
Thanh D, Prasath S, Le Minh H. A review on CT and X-Ray images denoising methods. Informatica. 2019;43:151-159.
Decuyper M, Maebe J, Van Holen R, Vandenberghe S. Artificial intelligence with deep learning in nuclear medicine and radiology. EJNMMI Physics. 2021;8(1):81.
Kim B, Han M, Shim H, Baek J. A performance comparison of convolutional neural network-based image denoising methods: the effect of loss functions on low-dose CT images. Med Phys. 2019;46(9):3906-3923.
Kulathilake KASH, Abdullah NA, Sabri AQM, Lai KW. A review on deep learning approaches for low-dose computed tomography restoration. Complex intell Systems. 2021.
Immonen E, Wong J, Nieminen M, et al. The use of deep learning towards dose optimization in low-dose computed tomography: a scoping review. Radiography. 2022;28(1):208-214.
Jing J, Xia W, Hou M, et al. Training low dose CT denoising network without high quality reference data. Phys Med Biol. 2022;67(8).
Bai T, Wang B, Nguyen D, Jiang S. Probabilistic self-learning framework for low-dose CT denoising. Med Phys. 2021;48(5):2258-2270.
Ulyanov D, Vedaldi A, Lempitsky V. Deep image prior. Int J Comput Vision. 2020;128:1867-1888.
Chen H, Zhang Y, Zhang W, et al. Low-dose CT denoising with convolutional neural network. In: 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017); 18-21 April 2017. 2017:143-146.
Chen H, Zhang Y, Kalra MK, et al. Low-Dose CT with a residual encoder-decoder convolutional neural network. IEEE Trans Med Imaging. 2017;36(12):2524-2535.
Kang E, Min J, Ye JC. A deep convolutional neural network using directional wavelets for low-dose X-ray CT reconstruction. Med Phys. 2017;44(10):e360-e375.
Yang W, Zhang H, Yang J, et al. Improving low-dose CT image using residual convolutional network. IEEE Access. 2017;5:24698-24705.
Wolterink JM, Leiner T, Viergever MA, Isgum I. Generative adversarial networks for noise reduction in low-dose CT. IEEE Trans Med Imaging. 2017;36(12):2536-2545.
Yang Q, Yan P, Zhang Y, et al. Low-Dose CT image denoising using a generative adversarial network with wasserstein distance and perceptual loss. IEEE Trans Med Imaging. 2017.
Tang C, Li J, Wang L, et al. Unpaired low-dose CT denoising network based on cycle-consistent generative adversarial network with prior image information. Comput Math Methods Med. 2019;2019:1-11.
Kang E, Koo HJ, Yang DH, Seo JB, Ye JC. Cycle-consistent adversarial denoising network for multiphase coronary CT angiography. Med Phys. 2019;46(2):550-562.
Yin Z, Xia K, He Z, Zhang J, Wang S, Zu B. Unpaired image denoising via wasserstein GAN in low-dose CT image with multi-perceptual loss and fidelity loss. Symmetry. 2021;13(1):126.
Han Y, Wu D, Kim K, Li Q. End-to-end deep learning for interior tomography with low-dose x-ray CT. Phys Med Biol. 2022;67(11):115001.
Akagi M, Nakamura Y, Higaki T, et al. Deep learning reconstruction improves image quality of abdominal ultra-high-resolution CT. Eur Radiol. 2019;29(11):6163-6171.
Chen L, Zheng L, Lian M, Luo S. A C-GAN denoising algorithm in projection domain for Micro-CT. Mol Cell Biomech. 2019;17:1-8.
Yao W, Chen L, Wu H, Zhao Q, Luo S. Micro-CT image denoising with an asymmetric perceptual convolutional network. Phys Med Biol. 2021:66.
Clark D, Badea C. Convolutional regularization methods for 4D, x-ray CT reconstruction. Vol 10948. SPIE; 2019.
Muller FM, Vanhove C, Vandeghinste B, Vandenberghe S. Performance evaluation of a micro-CT system for laboratory animal imaging with iterative reconstruction capabilities. Med Phys. 2022;49(5):3121-3133.
Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015. Springer; 2015.
Kingma D, Adam BJ. A method for stochastic optimization. In: International Conference on Learning Representations. 2014.
Lehtinen J, Munkberg J, Hasselgren J, et al. Noise2Noise: Learning Image Restoration without Clean Data. 2018.
Vandeghinste B, Vandenberghe S, Vanhove C, Staelens S, Van Holen R. Low-Dose Micro-CT imaging for vascular segmentation and analysis using sparse-view acquisitions. PLoS One. 2013;8(7):e68449.
Fleiss JL, Cohen J. The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educ Psychol Meas. 1973;33:613-619.
Fan L, Zhang F, Fan H, Zhang C. Brief review of image denoising techniques. Vis Comput Ind Biomed Art. 2019;2(1):7.
Wang C, Principe J. Training neural networks with additive noise in the desired signal. Neural Networks, IEEE Transactions on. 1999;10:1511-1517.
Wu D, Gong K, Kim K, Li Q. Consensus neural network for medical imaging denoising with only noisy training samples. ArXiv. 2019. abs/1906.03639.
Meganck JA, Liu B. Dosimetry in micro-computed tomography: a review of the measurement methods, impacts, and characterization of the quantum GX imaging system. Mol Imaging Biol. 2017;19(4):499-511.