Deep learning-based image quality improvement for low-dose computed tomography simulation in radiation therapy

Abstract. Low-dose computed tomography (CT) is desirable for treatment planning and simulation in radiation therapy. Multiple rescanning and replanning during the treatment course with a smaller amount of dose than a single conventional full-dose CT simulation is a crucial step in adaptive radiation therapy. We developed a machine learning-based method to improve image quality of low-dose CT for radiation therapy treatment simulation. We used a residual block concept and a self-attention strategy with a cycle-consistent adversarial network framework. A fully convolution neural network with residual blocks and attention gates (AGs) was used in the generator to enable end-to-end transformation. We have collected CT images from 30 patients treated with frameless brain stereotactic radiosurgery (SRS) for this study. These full-dose images were used to generate projection data, which were then added with noise to simulate the low-mAs scanning scenario. Low-dose CT images were reconstructed from this noise-contaminated projection data and were fed into our network along with the original full-dose CT images for training. The performance of our network was evaluated by quantitatively comparing the high-quality CT images generated by our method with the original full-dose images. When mAs is reduced to 0.5% of the original CT scan, the mean square error of the CT images obtained by our method is ∼1.6  %  , with respect to the original full-dose images. The proposed method successfully improved the noise, contract-to-noise ratio, and nonuniformity level to be close to those of full-dose CT images and outperforms a state-of-the-art iterative reconstruction method. Dosimetric studies show that the average differences of dose-volume histogram metrics are <0.1  Gy (p  >  0.05). These quantitative results strongly indicate that the denoised low-dose CT images using our method maintains image accuracy and quality and are accurate enough for dose calculation in current CT simulation of brain SRS treatment. We also demonstrate the great potential for low-dose CT in the process of simulation and treatment planning.

[1]  P. Nguyen,et al.  Radiation exposure from imaging tests: is there an increased cancer risk? , 2011, Expert review of cardiovascular therapy.

[2]  Lei Zhu,et al.  Pixel‐wise estimation of noise statistics on iterative CT reconstruction from a single scan , 2017, Medical physics.

[3]  Atul Padole,et al.  CT radiation dose and iterative reconstruction techniques. , 2015, AJR. American journal of roentgenology.

[4]  Yang Lei,et al.  Learning‐based CBCT correction using alternating random forest based on auto‐context model , 2018, Medical physics.

[5]  Mannudeep K. Kalra,et al.  Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN) , 2017, ArXiv.

[6]  Quanzheng Li,et al.  Iterative Low-Dose CT Reconstruction With Priors Trained by Artificial Neural Network , 2017, IEEE Transactions on Medical Imaging.

[7]  Nadeem Javaid,et al.  Deep learning model integrating features and novel classifiers fusion for brain tumor segmentation , 2019, Microscopy research and technique.

[8]  Wei Zhao,et al.  An indirect transmission measurement-based spectrum estimation method for computed tomography , 2015, Physics in medicine and biology.

[9]  Lei Zhu,et al.  Image-domain non-uniformity correction for cone-beam CT , 2017, 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017).

[10]  Ken D. Sauer,et al.  A Model-Based Image Reconstruction Algorithm With Simultaneous Beam Hardening Correction for X-Ray CT , 2015, IEEE Transactions on Computational Imaging.

[11]  Yang Lei,et al.  Dosimetric study on learning-based cone-beam CT correction in adaptive radiation therapy. , 2019, Medical dosimetry : official journal of the American Association of Medical Dosimetrists.

[12]  Tian Liu,et al.  MRI-based pseudo CT synthesis using anatomical signature and alternating random forest with iterative refinement model , 2018, Journal of medical imaging.

[13]  Tian Liu,et al.  MRI-based treatment planning for brain stereotactic radiosurgery: Dosimetric validation of a learning-based pseudo-CT generation method. , 2019, Medical dosimetry : official journal of the American Association of Medical Dosimetrists.

[14]  Feng Lin,et al.  Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network , 2017, IEEE Transactions on Medical Imaging.

[15]  Jeffrey A. Fessler,et al.  Statistical image reconstruction for polyenergetic X-ray computed tomography , 2002, IEEE Transactions on Medical Imaging.

[16]  Amjad Rehman,et al.  Mobile-Health Applications for the Efficient Delivery of Health Care Facility to People with Dementia (PwD) and Support to Their Carers: A Survey , 2019, BioMed research international.

[17]  Zahid Mehmood,et al.  Classification of acute lymphoblastic leukemia using deep learning , 2018, Microscopy research and technique.

[18]  Lei Zhu,et al.  Noise suppression for energy-resolved CT using similarity-based non-local filtration , 2016, SPIE Medical Imaging.

[19]  Steve B. Jiang,et al.  Low-dose CT reconstruction via edge-preserving total variation regularization. , 2010, Physics in medicine and biology.

[20]  Jong Chul Ye,et al.  A deep convolutional neural network using directional wavelets for low‐dose X‐ray CT reconstruction , 2016, Medical physics.

[21]  Ken D. Sauer,et al.  Model-Based Iterative Reconstruction for Dual-Energy X-Ray CT Using a Joint Quadratic Likelihood Model , 2014, IEEE Transactions on Medical Imaging.

[22]  Steve B. Jiang,et al.  GPU-based iterative cone-beam CT reconstruction using tight frame regularization , 2010, Physics in medicine and biology.

[23]  Zahid Mehmood,et al.  Plasmodium life cycle stage classification based quantification of malaria parasitaemia in thin blood smears , 2018, Microscopy research and technique.

[24]  Zahid Mehmood,et al.  Feature enhancement framework for brain tumor segmentation and classification , 2019, Microscopy research and technique.

[25]  Lei Zhu,et al.  Dual energy CT with one full scan and a second sparse-view scan using structure preserving iterative reconstruction (SPIR) , 2015, Physics in medicine and biology.

[26]  D. Brenner,et al.  Computed tomography--an increasing source of radiation exposure. , 2007, The New England journal of medicine.

[27]  Naveed Islam,et al.  An ensemble classification of exudates in color fundus images using an evolutionary algorithm based optimal features selection , 2019, Microscopy research and technique.

[28]  Maik Stille,et al.  Residual U-Net Convolutional Neural Network Architecture for Low-Dose CT Denoising , 2018, Current Directions in Biomedical Engineering.

[29]  Jun Zhou,et al.  Learning-based automatic segmentation of arteriovenous malformations on contrast CT images in brain stereotactic radiosurgery. , 2019, Medical physics.

[30]  Uwe Kruger,et al.  Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction , 2019, Nat. Mach. Intell..

[31]  E. Fishman,et al.  CT scan parameters and radiation dose: practical advice for radiologists. , 2013, Journal of the American College of Radiology : JACR.

[32]  Lei Zhu,et al.  Noise suppression for dual-energy CT via penalized weighted least-square optimization with similarity-based regularization. , 2016, Medical physics.

[33]  Lei Zhu,et al.  Accelerated barrier optimization compressed sensing (ABOCS) reconstruction for cone-beam CT: Phantom studies. , 2012, Medical physics.

[34]  Steve B. Jiang,et al.  Low-dose CT reconstruction via edge-preserving total variation regularization , 2010, Physics in medicine and biology.

[35]  W A Kalender,et al.  Dose reduction in CT by anatomically adapted tube current modulation. I. Simulation studies. , 1999, Medical physics.

[36]  Quanzheng Li,et al.  Low‐dose CT reconstruction using spatially encoded nonlocal penalty , 2017, Medical physics.

[37]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[38]  Arpita Agarwal,et al.  Evaluating the image quality of cone beam CT acquired during rotational delivery. , 2015, The British journal of radiology.

[39]  Yang Lei,et al.  A learning-based automatic segmentation and quantification method on left ventricle in gated myocardial perfusion SPECT imaging: A feasibility study , 2019, Journal of Nuclear Cardiology.

[40]  D. Brenner,et al.  Estimated risks of radiation-induced fatal cancer from pediatric CT. , 2001, AJR. American journal of roentgenology.

[41]  Eugene C Lin,et al.  Radiation risk from medical imaging. , 2010, Mayo Clinic proceedings.

[42]  Tian Liu,et al.  Automatic multiorgan segmentation in thorax CT images using U-net-GAN. , 2019, Medical physics.

[43]  C. McCollough,et al.  Radiation dose reduction in computed tomography: techniques and future perspective. , 2009, Imaging in medicine.

[44]  Bo Lu,et al.  A comprehensive study on decreasing the kilovoltage cone‐beam CT dose by reducing the projection number , 2010, Journal of applied clinical medical physics.