Low-contrast lesions are difficult to detect in noisy low-dose CT images. Improving CT image quality for this detection task has the potential to improve diagnostic accuracy and patient outcomes. In this work, we use tunable neural networks for CT image restoration with a hyperparameter to control the variance/bias tradeoff. We use clinical images from a super-high-resolution normal-dose CT scan to synthesize low-contrast low-dose CT images for supervised training of deep learning CT reconstruction models. Those models are trained using with multiple noise realizations so that variance and bias can be penalized separately. We use a training loss function with one hyperparameter called the denoising level, which controls the variance/bias tradeoff. Finally, we evaluate the CT image quality to find the optimal denoising level for low-contrast lesion detectability. We evaluate performance using a shallow neural network model classification model to represent a suboptimal image observer. Our results indicate that the optimal networks for low-contrast lesion detectability are those that prioritize bias reduction rather than mean-squared error, which demonstrates the potential clinical benefit of our proposed tunable neural networks.
Wide-coverage detector CT and ultra-high-resolution (UHR) detector CT are two important features for current cardiac imaging modalities. The former one enables the scanner to finish a whole heart image scan in one bed position; the latter one gives superior resolution in fine structures such as stenoses, calcifications, implemented stents, and small vessel boundaries. However, no commercially available scanner has both these features in one scanner as of today. Herein, we propose to use existing UHR-CT data to train a super resolution (SR) neural network and apply the network in a wide-coverage detector CT system. The purpose of the network is to enhance the system resolution performance and reduce the noise while maintaining its wide-coverage feature without additional hardware changes. Thirteen UHR-CT patient datasets and their simulated-normal-resolution pairs were used for training a 3D residual-block U-Net. The modulation transfer function (MTF) measured from Catphan phantom scans showed the proposed super-resolution aided deep learning-based reconstruction (SR-DLR) improved the MTF resolution by relative ~30% and ~10% as compared to filtered-back projection and model-based iterative reconstruction approaches. In real patient cases, the SR-DLR images show better noise texture and enhanced spatial resolution along with better aortic valve, stent, calcification, and soft tissue features as compared to other reconstruction approaches.
Wide-coverage detector CT and ultra-high-resolution (UHR) detector CT are two important features for current cardiac imaging modalities. The former one enables the scanner to finish a whole heart image scan in one bed position; the latter one gives superior resolution in fine structures such as stenoses, calcifications, implemented stents, and small vessel boundaries. However, no commercially available scanner has both these features in one scanner as of today. Herein, we propose to use existing UHR-CT data to train a super resolution (SR) neural network and apply the network in a widecoverage detector CT system. The purpose of the network is to enhance the system resolution performance and reduce the noise while maintaining its wide-coverage feature without additional hardware changes. Thirteen UHR-CT patient datasets and their simulated-normal-resolution pairs were used for training a 3D residual-block U-Net. The modulation transfer function (MTF) measured from Catphan phantom scans showed the proposed super-resolution aided deep learning-based reconstruction (SR-DLR) improved the MTF resolution by relative ~30% and ~10% as compared to filtered-back projection and model-based iterative reconstruction approaches. In real patient cases, the SR-DLR images show better noise texture and enhanced spatial resolution along with better aortic valve, stent, calcification, and soft tissue features as compared to other reconstruction approaches.
In conventional x-ray CT imaging, noise reduction is often applied on raw data to remove noise while improving reconstruction quality. Adaptive data filtering is one noise reduction method that suppresses data noise using a local smooth kernel. The design of the local kernel is important and can greatly affect the reconstruction quality. In this report we develop a deep learning convolutional neural network to help predict the local kernel automatically and adaptively to the data statistics. The proposed network is trained to directly generate kernel parameters and hence allow fast data filtering. We compare our method to the existing filtering method. The results shows that our deep learning based method is more efficient and robust over a variety of scan conditions.
Due to the wide variability of intra-patient respiratory motion patterns, traditional short-duration cine CT used in respiratory gated PET/CT may be insufficient to match the PET scan data, resulting in suboptimal attenuation correction that eventually compromises the PET quantitative accuracy. Thus, extending the duration of cine CT can be beneficial to address this data mismatch issue. In this work, we propose to use a long-duration cine CT for respiratory gated PET/CT, whose cine acquisition time is ten times longer than a traditional short-duration cine CT. We compare the proposed long-duration cine CT with the traditional short-duration cine CT through numerous phantom simulations with 11 respiratory traces measured during patient PET/CT scans. Experimental results show that, the long-duration cine CT reduces the motion mismatch between PET and CT by 41% and improves the overall reconstruction accuracy by 42% on average, as compared to the traditional short-duration cine CT. The long-duration cine CT also reduces artifacts in PET images caused by misalignment and mismatch between adjacent slices in phase-gated CT images. The improvement in motion matching between PET and CT by extending the cine duration depends on the patient, with potentially greater benefits for patients with irregular breathing patterns or larger diaphragm movements.
Low dose CT imaging is typically constrained to be diagnostic. However, there are applications for even lowerdose CT imaging, including image registration across multi-frame CT images and attenuation correction for PET/CT imaging. We define this as the ultra-low-dose (ULD) CT regime where the exposure level is a factor of 10 lower than current low-dose CT technique levels. In the ULD regime it is possible to use statistically-principled image reconstruction methods that make full use of the raw data information. Since most statistical based iterative reconstruction methods are based on the assumption of that post-log noise distribution is close to Poisson or Gaussian, our goal is to understand the statistical distribution of ULD CT data with different non-positivity correction methods, and to understand when iterative reconstruction methods may be effective in producing images that are useful for image registration or attenuation correction in PET/CT imaging. We first used phantom measurement and calibrated simulation to reveal how the noise distribution deviate from normal assumption under the ULD CT flux environment. In summary, our results indicate that there are three general regimes: (1) Diagnostic CT, where post-log data are well modeled by normal distribution. (2) Lowdose CT, where normal distribution remains a reasonable approximation and statistically-principled (post-log) methods that assume a normal distribution have an advantage. (3) An ULD regime that is photon-starved and the quadratic approximation is no longer effective. For instance, a total integral density of 4.8 (ideal pi for ~24 cm of water) for 120kVp, 0.5mAs of radiation source is the maximum pi value where a definitive maximum likelihood value could be found. This leads to fundamental limits in the estimation of ULD CT data when using a standard data processing stream
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.