Structurally-sensitive Multi-scale Deep Neural Network for Low-Dose CT Denoising

Asymmetric Convolution-based GAN Framework for Low-Dose CT Image Denoising

Noise reduction is essential to improve the diagnostic quality of low-dose CT (LDCT) images. In this regard, data-driven denoising methods based on generative adversarial networks (GAN) have shown promising results. However, custom designs with 2D convolution may not preserve the correlation of the local and global pixels, which results in the loss of high-frequency (edges/ boundaries of lesions) anatomical details. A recent state-of-the-art method demonstrates that using primitive GAN-based methods may introduce structural (shape) distortion. To address this issue, we develop a novel asymmetric convolution-based generator network (ACGNet), which is constructed by using one-dimensional (1D) asymmetric convolutions and a dynamic attention module (DAM). The 1D asymmetric convolutions (1 × 3 & 3 × 1) can intensify the representation power of square convolution kernels (3 × 3) in horizontal and vertical directions. Consequently, we integrated the highlighted low-level CT voxel details via purposed attention DAM with high-level CT-scan features. As a result, ACGNet efficiently preserves the local and global pixel relations in denoised LDCT images. Furthermore, we propose a novel neural structure preserving loss (NSPL) through which ACGNet learns the neighborhood structure of CT images, preventing structural (shape) distortion. In addition, the ACGNet can reconstruct the CT images with human-perceived quality via back-propagated gradients due to the feature-based NSPL loss. Finally, we include differential content loss in network optimization to restore high-frequency lesion boundaries. The proposed method outperforms many state-of-the-art methods on two publicly accessible datasets: the Mayo 2016 dataset (PSNR: 35.2015 dB, SSIM: 0.9560), and Low-dose CT image and projection dataset (PSNR: 35.2825 dB, SSIM: 0.9566).

TQGDNet: Coronary artery calcium deposit detection on computed tomography

Coronary artery disease (CAD) continues to be a leading global cause of cardiovascular related mortality. The scoring of coronary artery calcium (CAC) using computer tomography (CT) images is a diagnostic instrument for evaluating the risk of asymptomatic individuals prone to atherosclerotic cardiovascular disease. State-of-the-art automated CAC scoring methods rely on large annotated datasets to train convolutional neural network (CNN) models. However, these methods do not integrate features across different levels and layers of the CNN, particularly in the lower layers where important information regarding small calcium regions are present. In this study, we propose a new CNN model specifically designed to effectively capture features associated with small regions and their surrounding areas in low-contrast CT images. Our model integrates a specifically designed low-contrast detection module and two fusion modules focusing on the lower layers of the network to connect more deeper and wider neurons (or nodes) across multiple adjacent levels. Our first module, called ThrConvs, includes three convolution blocks tailored to detecting objects in images characterized by low contrast. Following this, two fusion modules are introduced: (i) Queen-fusion (Qf), which introduces a cross-scale feature method to fuse features from multiple adjacent levels and layers and, (ii) lower-layer Gather-and-Distribute (GD) module, which focuses on learning comprehensive features associated with small-sized calcium deposits and their surroundings. We demonstrate superior performance of our model using the public OrCaScore dataset, encompassing 269 calcium deposits, surpassing the capabilities of previous state-of-the-art works. We demonstrate the enhanced performance of our approach, achieving a notable 2.3-3.6 % improvement in mean Pixel Accuracy (mPA) on both the private Concord dataset and the public OrCaScore dataset, surpassing the capabilities of established detection methods.

Assessment of perceived realism in AI-generated synthetic spine fracture CT images

BackgroundDeep learning-based decision support systems require synthetic images generated by adversarial networks, which require clinical evaluation to ensure their quality.ObjectiveThe study evaluates perceived realism of high-dimension synthetic spine fracture CT images generated Progressive Growing Generative Adversarial Networks (PGGANs).Method: The study used 2820 spine fracture CT images from 456 patients to train an PGGAN model. The model synthesized images up to 512 × 512 pixels, and the realism of the generated images was assessed using Visual Turing Tests and Fracture Identification Test. Three spine surgeons evaluated the images, and clinical evaluation results were statistically analysed.Result: Spine surgeons have an average prediction accuracy of nearly 50% during clinical evaluations, indicating difficulty in distinguishing between real and generated images. The accuracy varies for different dimensions, with synthetic images being more realistic, especially in 512 × 512-dimension images. During FIT, among 16 generated images of each fracture type, 13-15 images were correctly identified, indicating images are more realistic and clearly depict fracture lines in 512 × 512 dimensions.ConclusionThe study reveals that AI-based PGGAN can generate realistic synthetic spine fracture CT images up to 512 × 512 pixels, making them difficult to distinguish from real images, and improving the automatic spine fracture type detection system.

Deep learning-based image domain reconstruction enhances image quality and pulmonary nodule detection in ultralow-dose CT with adaptive statistical iterative reconstruction-V

OBJECTIVES

To evaluate the image quality and lung nodule detectability of ultralow-dose CT (ULDCT) with adaptive statistical iterative reconstruction-V (ASiR-V) post-processed using a deep learning image reconstruction (DLIR)-based image domain compared to low-dose CT (LDCT) and ULDCT without DLIR.

MATERIALS AND METHODS

A total of 210 patients undergoing lung cancer screening underwent LDCT (mean ± SD, 0.81 ± 0.28 mSv) and ULDCT (0.17 ± 0.03 mSv) scans. ULDCT images were reconstructed with ASiR-V (ULDCT-ASiR-V) and post-processed using DLIR (ULDCT-DLIR). The quality of the three CT images was analyzed. Three radiologists detected and measured pulmonary nodules on all CT images, with LDCT results serving as references. Nodule conspicuity was assessed using a five-point Likert scale, followed by further statistical analyses.

RESULTS

A total of 463 nodules were detected using LDCT. The image noise of ULDCT-DLIR decreased by 60% compared to that of ULDCT-ASiR-V and was lower than that of LDCT (p < 0.001). The subjective image quality scores for ULDCT-DLIR (4.4 [4.1, 4.6]) were also higher than those for ULDCT-ASiR-V (3.6 [3.1, 3.9]) (p < 0.001). The overall nodule detection rates for ULDCT-ASiR-V and ULDCT-DLIR were 82.1% (380/463) and 87.0% (403/463), respectively (p < 0.001). The percentage difference between diameters > 1 mm was 2.9% (ULDCT-ASiR-V vs. LDCT) and 0.5% (ULDCT-DLIR vs. LDCT) (p = 0.009). Scores of nodule imaging sharpness on ULDCT-DLIR (4.0 ± 0.68) were significantly higher than those on ULDCT-ASiR-V (3.2 ± 0.50) (p < 0.001).

CONCLUSION

DLIR-based image domain improves image quality, nodule detection rate, nodule imaging sharpness, and nodule measurement accuracy of ASiR-V on ULDCT.

KEY POINTS

Question Deep learning post-processing is simple and cheap compared with raw data processing, but its performance is not clear on ultralow-dose CT. Findings Deep learning post-processing enhanced image quality and improved the nodule detection rate and accuracy of nodule measurement of ultralow-dose CT. Clinical relevance Deep learning post-processing improves the practicability of ultralow-dose CT and makes it possible for patients with less radiation exposure during lung cancer screening.

Self-supervised learning for CT image denoising and reconstruction: a review

This article reviews the self-supervised learning methods for CT image denoising and reconstruction. Currently, deep learning has become a dominant tool in medical imaging as well as computer vision. In particular, self-supervised learning approaches have attracted great attention as a technique for learning CT images without clean/noisy references. After briefly reviewing the fundamentals of CT image denoising and reconstruction, we examine the progress of deep learning in CT image denoising and reconstruction. Finally, we focus on the theoretical and methodological evolution of self-supervised learning for image denoising and reconstruction.

Integrating Learning-Based Priors With Physics-Based Models in Ultrasound Elasticity Reconstruction

Ultrasound elastography images, which enable quantitative visualization of tissue stiffness, can be reconstructed by solving an inverse problem. Classical model-based methods are usually formulated in terms of constrained optimization problems. To stabilize the elasticity reconstructions, regularization techniques, such as Tikhonov method, are used with the cost of promoting smoothness and blurriness in the reconstructed images. Thus, incorporating a suitable regularizer is essential for reducing the elasticity reconstruction artifacts, while finding the most suitable one is challenging. In this work, we present a new statistical representation of the physical imaging model, which incorporates effective signal-dependent colored noise modeling. Moreover, we develop a learning-based integrated statistical framework, which combines a physical model with learning-based priors. We use a dataset of simulated phantoms with various elasticity distributions and geometric patterns to train a denoising regularizer as the learning-based prior. We use fixed-point approaches and variants of gradient descent for solving the integrated optimization task following learning-based plug-and-play (PnP) prior and regularization by denoising (RED) paradigms. Finally, we evaluate the performance of the proposed approaches in terms of relative mean square error (RMSE) with nearly 20% improvement for both piecewise smooth simulated phantoms and experimental phantoms compared with the classical model-based methods and 12% improvement for both spatially varying breast-mimicking simulated phantoms and an experimental breast phantom, demonstrating the potential clinical relevance of our work. Moreover, the qualitative comparisons of reconstructed images demonstrate the robust performance of the proposed methods even for complex elasticity structures that might be encountered in clinical settings.

Deep-learning-based attenuation map generation in kidney single photon emission computed tomography

BACKGROUND

Accurate attenuation correction (AC) is vital in nuclear medicine, particularly for quantitative single-photon emission computed tomography/computed tomography (SPECT/CT) imaging. This study aimed to establish a CT-free quantification technology in kidney SPECT imaging using deep learning to generate synthetic attenuation maps (μ-maps) from SPECT data, thereby reducing radiation exposure and eliminating the need for CT scans.

RESULTS

A dataset of 1000 Tc-99m DTPA SPECT/CT scans was analyzed for training (n = 800), validation (n = 100), and testing (n = 100) using a modified 3D U-Net for deep learning. The study investigated the use of primary emission and scattering SPECT data, normalization methods, loss function optimization, and up-sampling techniques for optimal μ-map generation. The problem of checkerboard artifacts, unique to μ-map generation from SPECT signals, and the effects of iodine contrast media were evaluated. The addition of scattering SPECT to primary emission SPECT imaging, logarithmic maximum normalization, the combination of absolute difference loss (L1) and three times the absolute gradient difference loss (3 × LGDL), and the nearest-neighbor interpolation significantly enhanced AI performance in μ-map generation (p < 0.00001). Checkerboard artifacts were effectively eliminated using the nearest-neighbor interpolation technique. The developed AI algorithm produced μ-maps neutral to the presence of iodine contrast and showed negligible contrast effects on quantitative SPECT measurement, such as glomerular filtration rate (GFR). The potential reduction in radiation exposure by transitioning to AI-based CT-free SPECT imaging ranges from 45.3 to 78.8%.

CONCLUSION

The study successfully developed and optimized a deep learning algorithm for generating synthetic μ-maps in kidney SPECT images, demonstrating the potential to transition from conventional SPECT/CT to CT-free SPECT imaging for GFR measurement. This advancement represents a significant step towards enhancing patient safety and efficiency in nuclear medicine.

Assessing the Capacity of a Denoising Diffusion Probabilistic Model to Reproduce Spatial Context

Diffusion models have emerged as a popular family of deep generative models (DGMs). In the literature, it has been claimed that one class of diffusion models-denoising diffusion probabilistic models (DDPMs)-demonstrate superior image synthesis performance as compared to generative adversarial networks (GANs). To date, these claims have been evaluated using either ensemble-based methods designed for natural images, or conventional measures of image quality such as structural similarity. However, there remains an important need to understand the extent to which DDPMs can reliably learn medical imaging domain-relevant information, which is referred to as 'spatial context' in this work. To address this, a systematic assessment of the ability of DDPMs to learn spatial context relevant to medical imaging applications is reported for the first time. A key aspect of the studies is the use of stochastic context models (SCMs) to produce training data. In this way, the ability of the DDPMs to reliably reproduce spatial context can be quantitatively assessed by use of post-hoc image analyses. Error-rates in DDPM-generated ensembles are reported, and compared to those corresponding to other modern DGMs. The studies reveal new and important insights regarding the capacity of DDPMs to learn spatial context. Notably, the results demonstrate that DDPMs hold significant capacity for generating contextually correct images that are 'interpolated' between training samples, which may benefit data-augmentation tasks in ways that GANs cannot.

Dual stage MRI image restoration based on blind spot denoising and hybrid attention

BACKGROUND

Magnetic Resonance Imaging (MRI) is extensively utilized in clinical diagnostics and medical research, yet the imaging process is often compromised by noise interference. This noise arises from various sources, leading to a reduction in image quality and subsequently hindering the accurate interpretation of image details by clinicians. Traditional denoising methods typically assume that noise follows a Gaussian distribution, thereby neglecting the more complex noise types present in MRI images, such as Rician noise. As a result, denoising remains a challenging and practical task.

METHOD

The main research work of this paper focuses on modifying mask information based on a global mask mapper. The mask mapper samples all blind spot pixels on the denoised image and maps them to the same channel. By incorporating perceptual loss, it utilizes all available information to improve performance while avoiding identity mapping. During the denoising process, the model may mistakenly remove some useful information as noise, resulting in a loss of detail in the denoised image. To address this issue, we train a generative adversarial network (GAN) with adaptive hybrid attention to restore the detailed information in the denoised MRI images.

RESULT

The two-stage model NRAE shows an improvement of nearly 1.4 dB in PSNR and approximately 0.1 in SSIM on clinical datasets compared to other classic models. Specifically, compared to the baseline model, PSNR is increased by about 0.6 dB, and SSIM is only 0.015 lower. From a visual perspective, NRAE more effectively restores the details in the images, resulting in richer and clearer representation of image details.

CONCLUSION

We have developed a deep learning-based two-stage model to address noise issues in medical MRI images. This method not only successfully reduces noise signals but also effectively restores anatomical details. The current results indicate that this is a promising approach. In future work, we plan to replace the current denoising network with more advanced models to further enhance performance.

Development and validation of a noise insertion algorithm for photon-counting-detector CT

BACKGROUND

Inserting noise into existing patient projection data to simulate lower-radiation-dose exams has been frequently used in traditional energy-integrating-detector (EID)-CT to optimize radiation dose in clinical protocols and to generate paired images for training deep-learning-based reconstruction and noise reduction methods. Recent introduction of photon counting detector CT (PCD-CT) also requires such a method to accomplish these tasks. However, clinical PCD-CT scanners often restrict the users access to the raw count data, exporting only the preprocessed, log-normalized sinogram. Therefore, it remains a challenge to employ projection domain noise insertion algorithms on PCD-CT.

PURPOSE

To develop and validate a projection domain noise insertion algorithm for PCD-CT that does not require access to the raw count data.

MATERIALS AND METHODS

A projection-domain noise model developed originally for EID-CT was adapted for PCD-CT. This model requires, as input, a map of the incident number of photons at each detector pixel when no object is in the beam. To obtain the map of incident number of photons, air scans were acquired on a PCD-CT scanner, then the noise equivalent photon number (NEPN) was calculated from the variance in the log normalized projection data of each scan. Additional air scans were acquired at various mA settings to investigate the impact of pulse pileup on the linearity of NEPN measurement. To validate the noise insertion algorithm, Noise Power Spectra (NPS) were generated from a 30 cm water tank scan and used to compare the noise texture and noise level of measured and simulated half dose and quarter dose images. An anthropomorphic thorax phantom was scanned with automatic exposure control, and noise levels at different slice locations were compared between simulated and measured half dose and quarter dose images. Spectral correlation between energy thresholds T1 and T2, and energy bins, B1 and B2, was compared between simulated and measured data across a wide range of tube current. Additionally, noise insertion was performed on a clinical patient case for qualitative assessment.

RESULTS

The NPS generated from simulated low dose water tank images showed similar shape and amplitude to that generated from the measured low dose images, differing by a maximum of 5.0% for half dose (HD) T1 images, 6.3% for HD T2 images, 4.1% for quarter dose (QD) T1 images, and 6.1% for QD T2 images. Noise versus slice measurements of the lung phantom showed comparable results between measured and simulated low dose images, with root mean square percent errors of 5.9%, 5.4%, 5.0%, and 4.6% for QD T1, HD T1, QD T2, and HD T2, respectively. NEPN measurements in air were linear up until 112 mA, after which pulse pileup effects significantly distort the air scan NEPN profile. Spectral correlation between T1 and T2 in simulation agreed well with that in the measured data in typical dose ranges.

CONCLUSIONS

A projection-domain noise insertion algorithm was developed and validated for PCD-CT to synthesize low-dose images from existing scans. It can be used for optimizing scanning protocols and generating paired images for training deep-learning-based methods.

Weakly supervised low-dose computed tomography denoising based on generative adversarial networks

Background

Low-dose computed tomography (LDCT) is a diagnostic imaging technique designed to minimize radiation exposure to the patient. However, this reduction in radiation may compromise computed tomography (CT) image quality, adversely impacting clinical diagnoses. Various advanced LDCT methods have emerged to mitigate this challenge, relying on well-matched LDCT and normal-dose CT (NDCT) image pairs for training. Nevertheless, these methods often face difficulties in distinguishing image details from nonuniformly distributed noise, limiting their denoising efficacy. Additionally, acquiring suitably paired datasets in the medical domain poses challenges, further constraining their applicability. Hence, the objective of this study was to develop an innovative denoising framework for LDCT images employing unpaired data.

Methods

In this paper, we propose a LDCT denoising network (DNCNN) that alleviates the need for aligning LDCT and NDCT images. Our approach employs generative adversarial networks (GANs) to learn and model the noise present in LDCT images, establishing a mapping from the pseudo-LDCT to the actual NDCT domain without the need for paired CT images.

Results

Within the domain of weakly supervised methods, our proposed model exhibited superior objective metrics on the simulated dataset when compared to CycleGAN and selective kernel-based cycle-consistent GAN (SKFCycleGAN): the peak signal-to-noise ratio (PSNR) was 43.9441, the structural similarity index measure (SSIM) was 0.9660, and the visual information fidelity (VIF) was 0.7707. In the clinical dataset, we conducted a visual effect analysis by observing various tissues through different observation windows. Our proposed method achieved a no-reference structural sharpness (NRSS) value of 0.6171, which was closest to that of the NDCT images (NRSS =0.6049), demonstrating its superiority over other denoising techniques in preserving details, maintaining structural integrity, and enhancing edge contrast.

Conclusions

Through extensive experiments on both simulated and clinical datasets, we demonstrated the superior efficacy of our proposed method in terms of denoising quality and quantity. Our method exhibits superiority over both supervised techniques, including block-matching and 3D filtering (BM3D), residual encoder-decoder convolutional neural network (RED-CNN), and Wasserstein generative adversarial network-VGG (WGAN-VGG), and over weakly supervised approaches, including CycleGAN and SKFCycleGAN.

Subjective and objective image quality of low-dose CT images processed using a self-supervised denoising algorithm

This study aimed to assess the subjective and objective image quality of low-dose computed tomography (CT) images processed using a self-supervised denoising algorithm with deep learning. We trained the self-supervised denoising model using low-dose CT images of 40 patients and applied this model to CT images of another 30 patients. Image quality, in terms of noise and edge sharpness, was rated on a 5-point scale by two radiologists. The coefficient of variation, contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR) were calculated. The values for the self-supervised denoising model were compared with those for the original low-dose CT images and CT images processed using other conventional denoising algorithms (non-local means, block-matching and 3D filtering, and total variation minimization-based algorithms). The mean (standard deviation) scores of local and overall noise levels for the self-supervised denoising algorithm were 3.90 (0.40) and 3.93 (0.51), respectively, outperforming the original image and other algorithms. Similarly, the mean scores of local and overall edge sharpness for the self-supervised denoising algorithm were 3.90 (0.40) and 3.75 (0.47), respectively, surpassing the scores of the original image and other algorithms. The CNR and SNR for the self-supervised denoising algorithm were higher than those for the original images but slightly lower than those for the other algorithms. Our findings indicate the potential clinical applicability of the self-supervised denoising algorithm for low-dose CT images in clinical settings.

Texture-preserving low dose CT image denoising using Pearson divergence

Objective.The mean squared error (MSE), also known asL2loss, has been widely used as a loss function to optimize image denoising models due to its strong performance as a mean estimator of the Gaussian noise model. Recently, various low-dose computed tomography (LDCT) image denoising methods using deep learning combined with the MSE loss have been developed; however, this approach has been observed to suffer from the regression-to-the-mean problem, leading to over-smoothed edges and degradation of texture in the image.Approach.To overcome this issue, we propose a stochastic function in the loss function to improve the texture of the denoised CT images, rather than relying on complicated networks or feature space losses. The proposed loss function includes the MSE loss to learn the mean distribution and the Pearson divergence loss to learn feature textures. Specifically, the Pearson divergence loss is computed in an image space to measure the distance between two intensity measures of denoised low-dose and normal-dose CT images. The evaluation of the proposed model employs a novel approach of multi-metric quantitative analysis utilizing relative texture feature distance.Results.Our experimental results show that the proposed Pearson divergence loss leads to a significant improvement in texture compared to the conventional MSE loss and generative adversarial network (GAN), both qualitatively and quantitatively.Significance.Achieving consistent texture preservation in LDCT is a challenge in conventional GAN-type methods due to adversarial aspects aimed at minimizing noise while preserving texture. By incorporating the Pearson regularizer in the loss function, we can easily achieve a balance between two conflicting properties. Consistent high-quality CT images can significantly help clinicians in diagnoses and supporting researchers in the development of AI-diagnostic models.

FSformer: A combined frequency separation network and transformer for LDCT denoising

Low-dose computed tomography (LDCT) has been widely concerned in the field of medical imaging because of its low radiation hazard to humans. However, under low-dose radiation scenarios, a large amount of noise/artifacts are present in the reconstructed image, which reduces the clarity of the image and is not conducive to diagnosis. To improve the LDCT image quality, we proposed a combined frequency separation network and Transformer (FSformer) for LDCT denoising. Firstly, FSformer decomposes the LDCT images into low-frequency images and multi-layer high-frequency images by frequency separation blocks. Then, the low-frequency components are fused with the high-frequency components of different layers to remove the noise in the high-frequency components with the help of the potential texture of low-frequency parts. Next, the estimated noise images can be obtained by using Transformer stage in the frequency aggregation denoising block. Finally, they are fed into the reconstruction prediction block to obtain improved quality images. In addition, a compound loss function with frequency loss and Charbonnier loss is used to guide the training of the network. The performance of FSformer has been validated and evaluated on AAPM Mayo dataset, real Piglet dataset and clinical dataset. Compared with previous representative models in different architectures, FSformer achieves the optimal metrics with PSNR of 33.7714 dB and SSIM of 0.9254 on Mayo dataset, the testing time is 1.825 s. The experimental results show that FSformer is a state-of-the-art (SOTA) model with noise/artifact suppression and texture/organization preservation. Moreover, the model has certain robustness and can effectively improve LDCT image quality.

A systematic literature review: deep learning techniques for synthetic medical image generation and their applications in radiotherapy

The aim of this systematic review is to determine whether Deep Learning (DL) algorithms can provide a clinically feasible alternative to classic algorithms for synthetic Computer Tomography (sCT). The following categories are presented in this study: ∙ MR-based treatment planning and synthetic CT generation techniques. ∙ Generation of synthetic CT images based on Cone Beam CT images. ∙ Low-dose CT to High-dose CT generation. ∙ Attenuation correction for PET images. To perform appropriate database searches, we reviewed journal articles published between January 2018 and June 2023. Current methodology, study strategies, and results with relevant clinical applications were analyzed as we outlined the state-of-the-art of deep learning based approaches to inter-modality and intra-modality image synthesis. This was accomplished by contrasting the provided methodologies with traditional research approaches. The key contributions of each category were highlighted, specific challenges were identified, and accomplishments were summarized. As a final step, the statistics of all the cited works from various aspects were analyzed, which revealed that DL-based sCTs have achieved considerable popularity, while also showing the potential of this technology. In order to assess the clinical readiness of the presented methods, we examined the current status of DL-based sCT generation.

A Review of deep learning methods for denoising of medical low-dose CT images

To prevent patients from being exposed to excess of radiation in CT imaging, the most common solution is to decrease the radiation dose by reducing the X-ray, and thus the quality of the resulting low-dose CT images (LDCT) is degraded, as evidenced by more noise and streaking artifacts. Therefore, it is important to maintain high quality CT image while effectively reducing radiation dose. In recent years, with the rapid development of deep learning technology, deep learning-based LDCT denoising methods have become quite popular because of their data-driven and high-performance features to achieve excellent denoising results. However, to our knowledge, no relevant article has so far comprehensively introduced and reviewed advanced deep learning denoising methods such as Transformer structures in LDCT denoising tasks. Therefore, based on the literatures related to LDCT image denoising published from year 2016-2023, and in particular from 2020 to 2023, this study presents a systematic survey of current situation, and challenges and future research directions in LDCT image denoising field. Four types of denoising networks are classified according to the network structure: CNN-based, Encoder-Decoder-based, GAN-based, and Transformer-based denoising networks, and each type of denoising network is described and summarized from the perspectives of structural features and denoising performances. Representative deep-learning denoising methods for LDCT are experimentally compared and analyzed. The study results show that CNN-based denoising methods capture image details efficiently through multi-level convolution operation, demonstrating superior denoising effects and adaptivity. Encoder-decoder networks with MSE loss, achieve outstanding results in objective metrics. GANs based methods, employing innovative generators and discriminators, obtain denoised images that exhibit perceptually a closeness to NDCT. Transformer-based methods have potential for improving denoising performances due to their powerful capability in capturing global information. Challenges and opportunities for deep learning based LDCT denoising are analyzed, and future directions are also presented.

CT image denoising methods for image quality improvement and radiation dose reduction

With the ever-increasing use of computed tomography (CT), concerns about its radiation dose have become a significant public issue. To address the need for radiation dose reduction, CT denoising methods have been widely investigated and applied in low-dose CT images. Numerous noise reduction algorithms have emerged, such as iterative reconstruction and most recently, deep learning (DL)-based approaches. Given the rapid advancements in Artificial Intelligence techniques, we recognize the need for a comprehensive review that emphasizes the most recently developed methods. Hence, we have performed a thorough analysis of existing literature to provide such a review. Beyond directly comparing the performance, we focus on pivotal aspects, including model training, validation, testing, generalizability, vulnerability, and evaluation methods. This review is expected to raise awareness of the various facets involved in CT image denoising and the specific challenges in developing DL-based models.

GAN-based medical image small region forgery detection via a two-stage cascade framework

Using generative adversarial network (GAN) Goodfellow et al. (2014) for data enhancement of medical images is significantly helpful for many computer-aided diagnosis (CAD) tasks. A new GAN-based automated tampering attack, like CT-GAN Mirsky et al. (2019), has emerged. It can inject or remove lung cancer lesions to CT scans. Because the tampering region may even account for less than 1% of the original image, even state-of-the-art methods are challenging to detect the traces of such tampering. This paper proposes a two-stage cascade framework to detect GAN-based medical image small region forgery like CT-GAN. In the local detection stage, we train the detector network with small sub-images so that interference information in authentic regions will not affect the detector. We use depthwise separable convolution and residual networks to prevent the detector from over-fitting and enhance the ability to find forged regions through the attention mechanism. The detection results of all sub-images in the same image will be combined into a heatmap. In the global classification stage, using gray-level co-occurrence matrix (GLCM) can better extract features of the heatmap. Because the shape and size of the tampered region are uncertain, we use hyperplanes in an infinite-dimensional space for classification. Our method can classify whether a CT image has been tampered and locate the tampered position. Sufficient experiments show that our method can achieve excellent performance than the state-of-the-art detection methods.

Measurement of proximal contact of single crowns to assess interproximal relief: A pilot study

Background

It is common for dental technicians to adjust the proximal surface of adjacent teeth on casts when fabricating single crowns. However, whether the accuracy of the proximal contact is affected if this step is eliminated is unclear.

Objective

To evaluate the accuracy of the proximal contact of single crowns for mandibular first molars fabricated from four different restorative materials, without adjustment of the proximal surface of the adjacent teeth by the laboratory/dental technician.

Methods

This study was in vitro; all the clinical procedures were conducted on a dentoform. The mandibular first molar tooth on the dentoform was prepared using diamond burs and a high speed handpiece. Twenty single crowns were fabricated, five for each group (monolithic zirconia, lithium disilicate, metal ceramic, and cast gold). No proximal surface adjacent to the definitive crowns was adjusted for tight contact in the dental laboratory. Both the qualitative analyses, using dental floss and shimstock, and the quantitative analyses, using a stereo microscope, were performed to evaluate the accuracy of the proximal contact of the restoration with the adjacent teeth. In the quantitative analysis, one-way analysis of variance was used to compare mean values at a significance level of 0.05.

Results

In quantitative analysis, the differences between the proximal contact tightness of the four groups was not statistically significant (P = 0.802 for mesial contacts, P = 0.354 for distal contacts). In qualitative analysis, in most crowns, dental floss passed through the contact with tight resistance and only one film of shimstock could be inserted between the adjacent teeth and the restoration. However, one specimen from the cast gold crown had open contact.

Conclusions

Even without proximal surface adjustment of the adjacent teeth during the crown fabrication process, adequate proximal contact tightness between the restoration and adjacent teeth could be achieved.

MLNAN: Multi-level noise-aware network for low-dose CT imaging implemented with constrained cycle Wasserstein generative adversarial networks

Low-dose CT techniques attempt to minimize the radiation exposure of patients by estimating the high-resolution normal-dose CT images to reduce the risk of radiation-induced cancer. In recent years, many deep learning methods have been proposed to solve this problem by building a mapping function between low-dose CT images and their high-dose counterparts. However, most of these methods ignore the effect of different radiation doses on the final CT images, which results in large differences in the intensity of the noise observable in CT images. What'more, the noise intensity of low-dose CT images exists significantly differences under different medical devices manufacturers. In this paper, we propose a multi-level noise-aware network (MLNAN) implemented with constrained cycle Wasserstein generative adversarial networks to recovery the low-dose CT images under uncertain noise levels. Particularly, the noise-level classification is predicted and reused as a prior pattern in generator networks. Moreover, the discriminator network introduces noise-level determination. Under two dose-reduction strategies, experiments to evaluate the performance of proposed method are conducted on two datasets, including the simulated clinical AAPM challenge datasets and commercial CT datasets from United Imaging Healthcare (UIH). The experimental results illustrate the effectiveness of our proposed method in terms of noise suppression and structural detail preservation compared with several other deep-learning based methods. Ablation studies validate the effectiveness of the individual components regarding the afforded performance improvement. Further research for practical clinical applications and other medical modalities is required in future works.

Degradation Adaption Local-to-Global Transformer for Low-Dose CT Image Denoising

Computer tomography (CT) has played an essential role in the field of medical diagnosis. Considering the potential risk of exposing patients to X-ray radiations, low-dose CT (LDCT) images have been widely applied in the medical imaging field. Since reducing the radiation dose may result in increased noise and artifacts, methods that can eliminate the noise and artifacts in the LDCT image have drawn increasing attentions and produced impressive results over the past decades. However, recent proposed methods mostly suffer from noise remaining, over-smoothing structures, or false lesions derived from noise. To tackle these issues, we propose a novel degradation adaption local-to-global transformer (DALG-Transformer) for restoring the LDCT image. Specifically, the DALG-Transformer is built on self-attention modules which excel at modeling long-range information between image patch sequences. Meanwhile, an unsupervised degradation representation learning scheme is first developed in medical image processing to learn abstract degradation representations of the LDCT images, which can distinguish various degradations in the representation space rather than the pixel space. Then, we introduce a degradation-aware modulated convolution and gated mechanism into the building modules (i.e., multi-head attention and feed-forward network) of each Transformer block, which can bring in the complementary strength of convolution operation to emphasize on the spatially local context. The experimental results show that the DALG-Transformer can provide superior performance in noise removal, structure preservation, and false lesions elimination compared with five existing representative deep networks. The proposed networks may be readily applied to other image processing tasks including image reconstruction, image deblurring, and image super-resolution.

Improvement of diagnostic performance of hyperacute ischemic stroke in head CT using an image-based noise reduction technique with non-black-boxed process

PURPOSE

This study aims to investigate whether an image-based noise reduction (INR) technique with a conventional rule-based algorithm involving no black-boxed processes can outperform an existing hybrid-type iterative reconstruction (HIR) technique, when applied to brain CT images for diagnosis of early CT signs, which generally exhibit low-contrast lesions that are difficult to detect.

METHODS

The subjects comprised 27 patients having infarctions within 4.5 h of onset and 27 patients with no change in brain parenchyma. Images with thicknesses of 5 mm and 0.625 mm were reconstructed by HIR. Images with a thickness of 0.625 mm reconstructed by filter back projection (FBP) were processed by INR. The contrast-to-noise ratios (CNRs) were calculated between gray and white matters; lentiform nucleus and internal capsule; infarcted and non-infarcted areas. Two radiologists subjectively evaluated the presence of hyperdense artery signs (HASs) and infarctions and visually scored three properties regarding image quality (0.625-mm HIR images were excluded because of their notably worse noise appearances).

RESULTS

The CNRs of INR were significantly better than those of HIR with P < 0.001 for all the indicators. INR yielded significantly higher areas under the curve for both infarction and HAS detections than HIR (P < 0.001). Also, INR significantly improved the visual scores of all the three indicators.

CONCLUSION

The INR incorporating a simple and reproducible algorithm was more effective than HIR in detecting early CT signs and can be potentially applied to CT images from a large variety of CT systems.

Self-trained deep convolutional neural network for noise reduction in CT

Purpose

Supervised deep convolutional neural network (CNN)-based methods have been actively used in clinical CT to reduce image noise. The networks of these methods are typically trained using paired high- and low-quality data from a massive number of patients and/or phantom images. This training process is tedious, and the network trained under a given condition may not be generalizable to patient images acquired and reconstructed under different conditions. We propose a self-trained deep CNN (ST_CNN) method for noise reduction in CT that does not rely on pre-existing training datasets.

Approach

The ST_CNN training was accomplished using extensive data augmentation in the projection domain, and the inference was applied to the data itself. Specifically, multiple independent noise insertions were applied to the original patient projection data to generate multiple realizations of low-quality projection data. Then, rotation augmentation was adopted for both the original and low-quality projection data by applying the rotation angle directly on the projection data so that images were rotated at arbitrary angles without introducing additional bias. A large number of paired low- and high-quality images from the same patient were reconstructed and paired for training the ST_CNN model.

Results

No significant difference was found between the ST_CNN and conventional CNN models in terms of the peak signal-to-noise ratio and structural similarity index measure. The ST_CNN model outperformed the conventional CNN model in terms of noise texture and homogeneity in liver parenchyma as well as better subjective visualization of liver lesions. The ST_CNN may sacrifice the sharpness of vessels slightly compared to the conventional CNN model but without affecting the visibility of peripheral vessels or diagnosis of vascular pathology.

Conclusions

The proposed ST_CNN method trained from the data itself may achieve similar image quality in comparison with conventional deep CNN denoising methods pre-trained on external datasets.

Generative adversarial network with radiomic feature reproducibility analysis for computed tomography denoising

BACKGROUND

Most computed tomography (CT) denoising algorithms have been evaluated using image quality analysis (IQA) methods developed for natural image, which do not adequately capture the texture details in medical imaging. Radiomics is an emerging image analysis technique that extracts texture information to provide a more objective basis for medical imaging diagnostics, overcoming the subjective nature of traditional methods. By utilizing the difficulty of reproducing radiomics features under different imaging protocols, we can more accurately evaluate the performance of CT denoising algorithms.

METHOD

We introduced radiomic feature reproducibility analysis as an evaluation metric for a denoising algorithm. Also, we proposed a low-dose CT denoising method based on a generative adversarial network (GAN), which outperformed well-known CT denoising methods.

RESULTS

Although the proposed model produced excellent results visually, the traditional image assessment metrics such as peak signal-to-noise ratio and structural similarity failed to show distinctive performance differences between the proposed method and the conventional ones. However, radiomic feature reproducibility analysis provided a distinctive assessment of the CT denoising performance. Furthermore, radiomic feature reproducibility analysis allowed fine-tuning of the hyper-parameters of the GAN.

CONCLUSION

We demonstrated that the well-tuned GAN architecture outperforms the well-known CT denoising methods. Our study is the first to introduce radiomics reproducibility analysis as an evaluation metric for CT denoising. We look forward that the study may bridge the gap between traditional objective and subjective evaluations in the clinical medical imaging field.

Applications of Artificial Intelligence in the Radiology Roundtrip: Process Streamlining, Workflow Optimization, and Beyond

There are many impactful applications of artificial intelligence (AI) in the electronic radiology roundtrip and the patient's journey through the healthcare system that go beyond diagnostic applications. These tools have the potential to improve quality and safety, optimize workflow, increase efficiency, and increase patient satisfaction. In this article, we review the role of AI for process improvement and workflow enhancement which includes applications beginning from the time of order entry, scan acquisition, applications supporting the image interpretation task, and applications supporting tasks after image interpretation such as result communication. These non-diagnostic workflow and process optimization tasks are an important part of the arsenal of potential AI tools that can streamline day to day clinical practice and patient care.

Transfer learning framework for low-dose CT reconstruction based on marginal distribution adaptation in multiscale

BACKGROUND

With the increasing use of computed tomography (CT) in clinical practice, limiting CT radiation exposure to reduce potential cancer risks has become one of the important directions of medical imaging research. As the dose decreases, the reconstructed CT image will be severely degraded by projection noise.

PURPOSE

As an important method of image processing, supervised deep learning has been widely used in the restoration of low-dose CT (LDCT) in recent years. However, the normal-dose CT (NDCT) corresponding to a specific LDCT (it is regarded as the label of the LDCT, which is necessary for supervised learning) is very difficult to obtain so that the application of supervised learning methods in LDCT reconstruction is limited. It is necessary to construct a unsupervised deep learning framework for LDCT reconstruction that does not depend on paired LDCT-NDCT datasets.

METHODS

We presented an unsupervised learning framework for the transferring from the identity mapping to the low-dose reconstruction task, called marginal distribution adaptation in multiscale (MDAM). For NDCTs as source domain data, MDAM is an identity map with two parts: firstly, it establishes a dimensionality reduction mapping, which can obtain the same feature distribution from NDCTs and LDCTs; and then NDCTs is retrieved by reconstructing the image overview and details from the low-dimensional features. For the purpose of the feature transfer between source domain and target domain (LDCTs), we introduce the multiscale feature extraction in the MDAM, and then eliminate differences in probability distributions of these multiscale features between NDCTs and LDCTs through wavelet decomposition and domain adaptation learning.

RESULTS

Image quality evaluation metrics and subjective quality scores show that, as an unsupervised method, the performance of the MDAM approaches or even surpasses some state-of-the-art supervised methods. Especially, MDAM has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection.

CONCLUSIONS

We demonstrated that, the MDAM framework can reconstruct corresponding NDCTs from LDCTs with high accuracy, and without relying on any labeles. Moreover, it is more suitable for clinical application compared with supervised learning methods.

Deep image and feature prior algorithm based on U-ConformerNet structure

PURPOSE

The reconstruction performance of the deep image prior (DIP) approach is limited by the conventional convolutional layer structure and it is difficult to enhance its potential. In order to improve the quality of image reconstruction and suppress artifacts, we propose a DIP algorithm with better performance, and verify its superiority in the latest case.

METHODS

We construct a new U-ConformerNet structure as the DIP algorithm's network, replacing the traditional convolutional layer-based U-net structure, and introduce the 'lpips' deep network based feature distance regularization method. Our algorithm can switch between supervised and unsupervised modes at will to meet different needs.

RESULTS

The reconstruction was performed on the low dose CT dataset (LoDoPaB). Our algorithm attained a PSNR of more than 35 dB under unsupervised conditions, and the PSNR under the supervised condition is greater than 36 dB. Both of which are better than the performance of the DIP-TV. Furthermore, the accuracy of this method is positively connected with the quality of the a priori image with the help of deep networks. In terms of noise eradication and artifact suppression, the DIP algorithm with U-ConformerNet structure outperforms the standard DIP method based on convolutional structure.

CONCLUSIONS

It is known by experimental verification that, in unsupervised mode, the algorithm improves the output PSNR by at least 2-3 dB when compared to the DIP-TV algorithm (proposed in 2020). In supervised mode, our algorithm approaches that of the state-of-the-art end-to-end deep learning algorithms.

A total variation prior unrolling approach for computed tomography reconstruction

BACKGROUND

With the rapid development of deep learning technology, deep neural networks can effectively enhance the performance of computed tomography (CT) reconstructions. One kind of commonly used method to construct CT reconstruction networks is to unroll the conventional iterative reconstruction (IR) methods to convolutional neural networks (CNNs). However, most unrolling methods primarily unroll the fidelity term of IR methods to CNNs, without unrolling the prior terms. The prior terms are always directly replaced by neural networks.

PURPOSE

In conventional IR methods, the prior terms play a vital role in improving the visual quality of reconstructed images. Unrolling the hand-crafted prior terms to CNNs may provide a more specialized unrolling approach to further improve the performance of CT reconstruction. In this work, a primal-dual network (PD-Net) was proposed by unrolling both the data fidelity term and the total variation (TV) prior term, which effectively preserves the image edges and textures in the reconstructed images.

METHODS

By further deriving the Chambolle-Pock (CP) algorithm instance for CT reconstruction, we discovered that the TV prior updates the reconstructed images with its divergences in each iteration of the solution process. Based on this discovery, CNNs were applied to yield the divergences of the feature maps for the reconstructed image generated in each iteration. Additionally, a loss function was applied to the predicted divergences of the reconstructed image to guarantee that the CNNs' results were the divergences of the corresponding feature maps in the iteration. In this manner, the proposed CNNs seem to play the same roles in the PD-Net as the TV prior in the IR methods. Thus, the TV prior in the CP algorithm instance can be directly unrolled to CNNs.

RESULTS

The datasets from the Low-Dose CT Image and Projection Data and the Piglet dataset were employed to assess the effectiveness of our proposed PD-Net. Compared with conventional CT reconstruction methods, our proposed method effectively preserves the structural and textural information in reference to ground truth.

CONCLUSIONS

The experimental results show that our proposed PD-Net framework is feasible for the implementation of CT reconstruction tasks. Owing to the promising results yielded by our proposed neural network, this study is intended to inspire further development of unrolling approaches by enabling the direct unrolling of hand-crafted prior terms to CNNs.

Image Turing test and its applications on synthetic chest radiographs by using the progressive growing generative adversarial network

The generative adversarial network (GAN) is a promising deep learning method for generating images. We evaluated the generation of highly realistic and high-resolution chest radiographs (CXRs) using progressive growing GAN (PGGAN). We trained two PGGAN models using normal and abnormal CXRs, solely relying on normal CXRs to demonstrate the quality of synthetic CXRs that were 1000 × 1000 pixels in size. Image Turing tests were evaluated by six radiologists in a binary fashion using two independent validation sets to judge the authenticity of each CXR, with a mean accuracy of 67.42% and 69.92% for the first and second trials, respectively. Inter-reader agreements were poor for the first (κ = 0.10) and second (κ = 0.14) Turing tests. Additionally, a convolutional neural network (CNN) was used to classify normal or abnormal CXR using only real images and/or synthetic images mixed datasets. The accuracy of the CNN model trained using a mixed dataset of synthetic and real data was 93.3%, compared to 91.0% for the model built using only the real data. PGGAN was able to generate CXRs that were identical to real CXRs, and this showed promise to overcome imbalances between classes in CNN training.

Opportunistic screening for low bone density using abdominopelvic computed tomography scans

BACKGROUND

While low bone density is a major burden on US health system, current osteoporosis screening guidelines by the US Preventive Services Task Force are limited to women aged ≥65 and all postmenopausal women with certain risk factors. Even within recommended screening groups, actual screening rates are low (<26%) and vary across socioeconomic groups. The proposed model can opportunistically screen patients using abdominal CT studies for low bone density who may otherwise go undiagnosed.

PURPOSE

To develop an artificial intelligence (AI) model for opportunistic screening of low bone density using both contrast and non-contrast abdominopelvic computed tomography (CT) exams, for the purpose of referral to traditional bone health management, which typically begins with dual energy X-ray absorptiometry (DXA).

METHODS

We collected 6083 contrast-enhanced CT imaging exams paired with DXA exams within ±6 months documented between May 2015 and August 2021 in a single institution with four major healthcare practice regions. Our fusion AI pipeline receives the coronal and axial plane images of a contrast enhanced abdominopelvic CT exam and basic patient demographics (age, gender, body cross section lengths) to predict risk of low bone mass. The models were trained on lumbar spine T-scores from DXA exams and tested on multi-site imaging exams. The model was again tested in a prospective group (N = 344) contrast-enhanced and non-contrast-enhanced studies.

RESULTS

The models were evaluated on the same test set (1208 exams)-(1) Baseline model using demographic factors from electronic medical records (EMR) - 0.7 area under the curve of receiver operator characteristic (AUROC); Imaging based models: (2) axial view - 0.83 AUROC; (3) coronal view- 0.83 AUROC; (4) Fusion model-Imaging + demographic factors - 0.86 AUROC. The prospective test yielded one missed positive DXA case with a hip prosthesis among 23 positive contrast-enhanced CT exams and 0% false positive rate for non-contrast studies. Both positive cases among non-contrast enhanced CT exams were successfully detected. While only about 8% patients from prospective study received a DXA exam within 2 years, about 30% were detected with low bone mass by the fusion model, highlighting the need for opportunistic screening.

CONCLUSIONS

The fusion model, which combines two planes of CT images and EMRs data, outperformed individual models and provided a high, robust diagnostic performance for opportunistic screening of low bone density using contrast and non-contrast CT exams. This model could potentially improve bone health risk assessment with no additional cost. The model's handling of metal implants is an ongoing effort.

A nonconvex model-based combined geometric calibration scheme for micro cone-beam CT with irregular trajectories

BACKGROUND

Many dedicated cone-beam CT (CBCT) systems have irregular scanning trajectories. Compared with the standard CBCT calibration, accurate calibration for CBCT systems with irregular trajectories is a more complex task, since the geometric parameters for each scanning view are variable. Most of the existing calibration methods assume that the intrinsic geometric relationship of the fiducials in the phantom is precisely known, and rarely delve deeper into the issue of whether the phantom accuracy is adapted to the calibration model.

PURPOSE

A high-precision phantom and a highly robust calibration model are interdependent and mutually supportive, and they are both important for calibration accuracy, especially for the high-resolution CBCT. Therefore, we propose a calibration scheme that considers both accurate phantom measurement and robust geometric calibration.

METHODS

Our proposed scheme consists of two parts: (1) introducing a measurement model to acquire the accurate intrinsic geometric relationship of the fiducials in the phantom; (2) developing a highly noise-robust nonconvex model-based calibration method. The measurement model in the first part is achieved by extending our previous high-precision geometric calibration model suitable for CBCT with circular trajectories. In the second part, a novel iterative method with optimization constraints based on a back-projection model is developed to solve the geometric parameters of each view.

RESULTS

The simulations and real experiments show that the measurement errors of the fiducial ball bearings (BBs) are within the subpixel level. With the help of the geometric relationship of the BBs obtained by our measurement method, the classic calibration method can achieve good calibration based on far fewer BBs. All metrics obtained in simulated experiments as well as in real experiments on Micro CT systems with resolutions of 9 and 4.5 μm show that the proposed calibration method has higher calibration accuracy than the competing classic method. It is particularly worth noting that although our measurement model proves to be very accurate, the classic calibration method based on this measurement model can only achieve good calibration results when the resolution of the measurement system is close to that of the system to be calibrated, but our calibration scheme enables high-accuracy calibration even when the resolution of the system to be calibrated is twice that of the measurement system.

CONCLUSIONS

The proposed combined geometrical calibration scheme does not rely on a phantom with an intricate pattern of fiducials, so it is applicable in Micro CT with high resolution. The two parts of the scheme, the "measurement model" and the "calibration model," prove to be of high accuracy. The combination of these two models can effectively improve the calibration accuracy, especially in some extreme cases.

Deep learning-based attenuation map generation with simultaneously reconstructed PET activity and attenuation and low-dose application

Objective. In PET/CT imaging, CT is used for positron emission tomography (PET) attenuation correction (AC). CT artifacts or misalignment between PET and CT can cause AC artifacts and quantification errors in PET. Simultaneous reconstruction (MLAA) of PET activity (λ-MLAA) and attenuation (μ-MLAA) maps was proposed to solve those issues using the time-of-flight PET raw data only. However,λ-MLAA still suffers from quantification error as compared to reconstruction using the gold-standard CT-based attenuation map (μ-CT). Recently, a deep learning (DL)-based framework was proposed to improve MLAA by predictingμ-DL fromλ-MLAA andμ-MLAA using an image domain loss function (IM-loss). However, IM-loss does not directly measure the AC errors according to the PET attenuation physics. Our preliminary studies showed that an additional physics-based loss function can lead to more accurate PET AC. The main objective of this study is to optimize the attenuation map generation framework for clinical full-dose¹⁸F-FDG studies. We also investigate the effectiveness of the optimized network on predicting attenuation maps for synthetic low-dose oncological PET studies.Approach. We optimized the proposed DL framework by applying different preprocessing steps and hyperparameter optimization, including patch size, weights of the loss terms and number of angles in the projection-domain loss term. The optimization was performed based on 100 skull-to-toe¹⁸F-FDG PET/CT scans with minimal misalignment. The optimized framework was further evaluated on 85 clinical full-dose neck-to-thigh¹⁸F-FDG cancer datasets as well as synthetic low-dose studies with only 10% of the full-dose raw data.Main results. Clinical evaluation of tumor quantification as well as physics-based figure-of-merit metric evaluation validated the promising performance of our proposed method. For both full-dose and low-dose studies, the proposed framework achieved <1% error in tumor standardized uptake value measures.Significance. It is of great clinical interest to achieve CT-less PET reconstruction, especially for low-dose PET studies.

Multislice input for 2D and 3D residual convolutional neural network noise reduction in CT

Purpose

Deep convolutional neural network (CNN)-based methods are increasingly used for reducing image noise in computed tomography (CT). Current attempts at CNN denoising are based on 2D or 3D CNN models with a single- or multiple-slice input. Our work aims to investigate if the multiple-slice input improves the denoising performance compared with the single-slice input and if a 3D network architecture is better than a 2D version at utilizing the multislice input.

Approach

Two categories of network architectures can be used for the multislice input. First, multislice images can be stacked channel-wise as the multichannel input to a 2D CNN model. Second, multislice images can be employed as the 3D volumetric input to a 3D CNN model, in which the 3D convolution layers are adopted. We make performance comparisons among 2D CNN models with one, three, and seven input slices and two versions of 3D CNN models with seven input slices and one or three output slices. Evaluation was performed on liver CT images using three quantitative metrics with full-dose images as reference. Visual assessment was made by an experienced radiologist.

Results

When the input channels of the 2D CNN model increases from one to three to seven, a trend of improved performance was observed. Comparing the three models with the seven-slice input, the 3D CNN model with a one-slice output outperforms the other models in terms of noise texture and homogeneity in liver parenchyma as well as subjective visualization of vessels.

Conclusions

We conclude the that multislice input is an effective strategy for improving performance for 2D deep CNN denoising models. The pure 3D CNN model tends to have a better performance than the other models in terms of continuity across axial slices, but the difference was not significant compared with the 2D CNN model with the same number of slices as the input.

Low-dose CT denoising with a high-level feature refinement and dynamic convolution network

BACKGROUND

Since the potential health risks of the radiation generated by computer tomography (CT), concerns have been expressed on reducing the radiation dose. However, low-dose CT (LDCT) images contain complex noise and artifacts, bringing uncertainty to medical diagnosis.

PURPOSE

Existing deep learning (DL)-based denoising methods are difficult to fully exploit hierarchical features of different levels, limiting the effect of denoising. Moreover, the standard convolution kernel is parameter sharing and cannot be adjusted dynamically with input change. This paper proposes an LDCT denoising network using high-level feature refinement and multiscale dynamic convolution to mitigate these problems.

METHODS

The dual network structure proposed in this paper consists of the feature refinement network (FRN) and the dynamic perception network (DPN). The FDN extracts features of different levels through residual dense connections. The high-level hierarchical information is transmitted to DPN to improve the low-level representations. In DPN, the two networks' features are fused by local channel attention (LCA) to assign weights in different regions and handle CT images' delicate tissues better. Then, the dynamic dilated convolution (DDC) with multibranch and multiscale receptive fields is proposed to enhance the expression and processing ability of the denoising network. The experiments were trained and tested on the dataset "NIH-AAPM-Mayo Clinic Low-Dose CT Grand Challenge," consisting of 10 anonymous patients with normal-dose abdominal CT and LDCT at 25% dose. In addition, external validation was performed on the dataset "Low Dose CT Image and Projection Data," which included 300 chest CT images at 10% dose and 300 head CT images at 25% dose.

RESULTS

Proposed method compared with seven mainstream LDCT denoising algorithms. On the Mayo dataset, achieved peak signal-to-noise ratio (PSNR): 46.3526 dB (95% CI: 46.0121-46.6931 dB) and structural similarity (SSIM): 0.9844 (95% CI: 0.9834-0.9854). Compared with LDCT, the average increase was 3.4159 dB and 0.0239, respectively. The results are relatively optimal and statistically significant compared with other methods. In external verification, our algorithm can cope well with ultra-low-dose chest CT images at 10% dose and obtain PSNR: 28.6130 (95% CI: 28.1680-29.0580 dB) and SSIM: 0.7201 (95% CI: 0.7101-0.7301). Compared with LDCT, PSNR/SSIM is increased by 3.6536dB and 0.2132, respectively. In addition, the quality of LDCT can also be improved in head CT denoising.

CONCLUSIONS

This paper proposes a DL-based LDCT denoising algorithm, which utilizes high-level features and multiscale dynamic convolution to optimize the network's denoising effect. This method can realize speedy denoising and performs well in noise suppression and detail preservation, which can be helpful for the diagnosis of LDCT.

Class-Aware Adversarial Transformers for Medical Image Segmentation

Transformers have made remarkable progress towards modeling long-range dependencies within the medical image analysis domain. However, current transformer-based models suffer from several disadvantages: (1) existing methods fail to capture the important features of the images due to the naive tokenization scheme; (2) the models suffer from information loss because they only consider single-scale feature representations; and (3) the segmentation label maps generated by the models are not accurate enough without considering rich semantic contexts and anatomical textures. In this work, we present CASTformer, a novel type of adversarial transformers, for 2D medical image segmentation. First, we take advantage of the pyramid structure to construct multi-scale representations and handle multi-scale variations. We then design a novel class-aware transformer module to better learn the discriminative regions of objects with semantic structures. Lastly, we utilize an adversarial training strategy that boosts segmentation accuracy and correspondingly allows a transformer-based discriminator to capture high-level semantically correlated contents and low-level anatomical features. Our experiments demonstrate that CASTformer dramatically outperforms previous state-of-the-art transformer-based approaches on three benchmarks, obtaining 2.54%-5.88% absolute improvements in Dice over previous models. Further qualitative experiments provide a more detailed picture of the model's inner workings, shed light on the challenges in improved transparency, and demonstrate that transfer learning can greatly improve performance and reduce the size of medical image datasets in training, making CASTformer a strong starting point for downstream medical image analysis tasks.

A dual-domain neural network based on sinogram synthesis for sparse-view CT reconstruction

OBJECTIVE

The dual-domain deep learning-based reconstruction techniques have enjoyed many successful applications in the field of medical image reconstruction. Applying the analytical reconstruction based operator to transfer the data from the projection domain to the image domain, the dual-domain techniques may suffer from the insufficient suppression or removal of streak artifacts in areas with the missing view data, when addressing the sparse-view reconstruction problems. In this work, to overcome this problem, an intelligent sinogram synthesis based back-projection network (iSSBP-Net) was proposed for sparse-view computed tomography (CT) reconstruction. In the iSSBP-Net method, a convolutional neural network (CNN) was involved in the dual-domain method to inpaint the missing view data in the sinogram before CT reconstruction.

METHODS

The proposed iSSBP-Net method fused a sinogram synthesis sub-network (SS-Net), a sinogram filter sub-network (SF-Net), a back-projection layer, and a post-CNN into an end-to-end network. Firstly, to inpaint the missing view data, the SS-Net employed a CNN to synthesize the full-view sinogram in the projection domain. Secondly, to improve the visual quality of the sparse-view CT images, the synthesized sinogram was filtered by a CNN. Thirdly, the filtered sinogram was brought into the image domain through the back-projection layer. Finally, to yield images of high visual sensitivity, the post-CNN was applied to restore the desired images from the outputs of the back-projection layer.

RESULTS

The numerical experiments demonstrate that the proposed iSSBP-Net is superior to all competing algorithms under different scanning condintions for sparse-view CT reconstruction. Compared to the competing algorithms, the proposed iSSBP-Net method improved the peak signal-to-noise ratio of the reconstructed images about 1.21 dB, 0.26 dB, 0.01 dB, and 0.37 dB under the scanning conditions of 360, 180, 90, and 60 views, respectively.

CONCLUSION

The promising reconstruction results indicate that involving the SS-Net in the dual-domain method is could be an effective manner to suppress or remove the streak artifacts in sparse-view CT images. Due to the promising results reconstructed by the iSSBP-Net method, this study is intended to inspire the further development of sparse-view CT reconstruction by involving a SS-Net in the dual-domain method.

Edge feature extraction-based dual CNN for LDCT denoising

In low-dose computed tomography (LDCT) denoising tasks, it is often difficult to balance edge/detail preservation and noise/artifact reduction. To solve this problem, we propose a dual convolutional neural network (CNN) based on edge feature extraction (Ed-DuCNN) for LDCT. Ed-DuCNN consists of two branches. One branch is the edge feature extraction subnet (Edge_Net) that can fully extract the edge details in the image. The other branch is the feature fusion subnet (Fusion_Net) that introduces an attention mechanism to fuse edge features and noisy image features. Specifically, first, shallow edge-specific detail features are extracted by trainable Sobel convolutional blocks and then are integrated into Edge_Net together with the LDCT images to obtain deep edge detail features. Finally, the input image, shallow edge detail, and deep edge detail features are fused in Fusion_Net to generate the final denoised image. The experimental results show that the proposed Ed-DuCNN can achieve competitive performance in terms of quantitative metrics and visual perceptual quality compared with that of state-of-the-art methods.

Wavelet subband-specific learning for low-dose computed tomography denoising

Deep neural networks have shown great improvements in low-dose computed tomography (CT) denoising. Early algorithms were primarily optimized to obtain an accurate image with low distortion between the denoised image and reference full-dose image at the cost of yielding an overly smoothed unrealistic CT image. Recent research has sought to preserve the fine details of denoised images with high perceptual quality, which has been accompanied by a decrease in objective quality due to a trade-off between perceptual quality and distortion. We pursue a network that can generate accurate and realistic CT images with high objective and perceptual quality within one network, achieving a better perception-distortion trade-off. To achieve this goal, we propose a stationary wavelet transform-assisted network employing the characteristics of high- and low-frequency domains of the wavelet transform and frequency subband-specific losses defined in the wavelet domain. We first introduce a stationary wavelet transform for the network training procedure. Then, we train the network using objective loss functions defined for high- and low-frequency domains to enhance the objective quality of the denoised CT image. With this network design, we train the network again after replacing the objective loss functions with perceptual loss functions in high- and low-frequency domains. As a result, we acquired denoised CT images with high perceptual quality using this strategy while minimizing the objective quality loss. We evaluated our algorithms on the phantom and clinical images, and the quantitative and qualitative results indicate that ours outperform the existing state-of-the-art algorithms in terms of objective and perceptual quality.

Three-dimensional conditional generative adversarial network-based virtual thin-slice technique for the morphological evaluation of the spine

Virtual thin-slice (VTS) technique is a generative adversarial network-based algorithm that can generate virtual 1-mm-thick CT images from images of 3–10-mm thickness. We evaluated the performance of VTS technique for assessment of the spine. VTS was applied to 4-mm-thick CT images of 73 patients, and the visibility of intervertebral spaces was evaluated on the 4-mm-thick and VTS images. The heights of vertebrae measured on sagittal images reconstructed from the 4-mm-thick images and VTS images were compared with those measured on images reconstructed from 1-mm-thick images. Diagnostic performance for the detection of compression fractures was also compared. The intervertebral spaces were significantly more visible on the VTS images than on the 4-mm-thick images (P < 0.001). The absolute value of the measured difference in mean vertebral height between the VTS and 1-mm-thick images was smaller than that between the 4-mm-thick and 1-mm-thick images (P < 0.01–0.54). The diagnostic performance of the VTS images for detecting compression fracture was significantly lower than that of the 4-mm-thick images for one reader (P = 0.02). VTS technique enabled the identification of each vertebral body, and enabled accurate measurement of vertebral height. However, this technique is not suitable for diagnosing compression fractures.

Nuclear-medicine probes: Where we are and where we are going

Nuclear medicine probes turned into the key for the identification and precise location of sentinel lymph nodes and other occult lesions (i.e., tumors) by using the systemic administration of radiotracers. Intraoperative nuclear probes are key in the surgical management of some malignancies as well as in the determination of positive surgical margins, thus reducing the extent and potential surgery morbidity. Depending on their application, nuclear probes are classified into two main categories, namely, counting and imaging. Although counting probes present a simple design, are handheld (to be moved rapidly), and provide only acoustic signals when detecting radiation, imaging probes, also known as cameras, are more hardware-complex and also able to provide images but at the cost of an increased intervention time as displacing the camera has to be done slowly. This review article begins with an introductory section to highlight the relevance of nuclear-based probes and their components as well as the main differences between ionization- (semiconductor) and scintillation-based probes. Then, the most significant performance parameters of the probe are reviewed (i.e., sensitivity, contrast, count rate capabilities, shielding, energy, and spatial resolution), as well as the different types of probes based on the target radiation nature, namely: gamma (γ), beta (β) (positron and electron), and Cherenkov. Various available intraoperative nuclear probes are finally compared in terms of performance to discuss the state-of-the-art of nuclear medicine probes. The manuscript concludes by discussing the ideal probe design and the aspects to be considered when selecting nuclear-medicine probes.

Patch-based artifact reduction for three-dimensional volume projection data of sparse-view micro-computed tomography

Micro-computed tomography (micro-CT) enables the non-destructive acquisition of three-dimensional (3D) morphological structures at the micrometer scale. Although it is expected to be used in pathology and histology to analyze the 3D microstructure of tissues, micro-CT imaging of tissue specimens requires a long scan time. A high-speed imaging method, sparse-view CT, can reduce the total scan time and radiation dose; however, it causes severe streak artifacts on tomographic images reconstructed with analytical algorithms due to insufficient sampling. In this paper, we propose an artifact reduction method for 3D volume projection data from sparse-view micro-CT. Specifically, we developed a patch-based lightweight fully convolutional network to estimate full-view 3D volume projection data from sparse-view 3D volume projection data. We evaluated the effectiveness of the proposed method using physically acquired datasets. The qualitative and quantitative results showed that the proposed method achieved high estimation accuracy and suppressed streak artifacts in the reconstructed images. In addition, we confirmed that the proposed method requires both short training and prediction times. Our study demonstrates that the proposed method has great potential for artifact reduction for 3D volume projection data under sparse-view conditions.

SR-CycleGAN: super-resolution of clinical CT to micro-CT level with multi-modality super-resolution loss

Purpose: We propose a super-resolution (SR) method, named SR-CycleGAN, for SR of clinical computed tomography (CT) images to the micro-focus x-ray CT CT ( μ CT ) level. Due to the resolution limitations of clinical CT (about 500 × 500 × 500    μ m 3 / voxel ), it is challenging to obtain enough pathological information. On the other hand, μ CT scanning allows the imaging of lung specimens with significantly higher resolution (about 50 × 50 × 50    μ m 3 / voxel or higher), which allows us to obtain and analyze detailed anatomical information. As a way to obtain detailed information such as cancer invasion and bronchioles from preoperative clinical CT images of lung cancer patients, the SR of clinical CT images to the μ CT level is desired. Approach: Typical SR methods require aligned pairs of low-resolution (LR) and high-resolution images for training, but it is infeasible to obtain precisely aligned paired clinical CT and μ CT images. To solve this problem, we propose an unpaired SR approach that can perform SR on clinical CT to the μ CT level. We modify a conventional image-to-image translation network named CycleGAN to an inter-modality translation network named SR-CycleGAN. The modifications consist of three parts: (1) an innovative loss function named multi-modality super-resolution loss, (2) optimized SR network structures for enlarging the input LR image to2 k -times by width and height to obtain the SR output, and (3) sub-pixel shuffling layers for reducing computing time. Results: Experimental results demonstrated that our method successfully performed SR of lung clinical CT images. SSIM and PSNR scores of our method were 0.54 and 17.71, higher than the conventional CycleGAN's scores of 0.05 and 13.64, respectively. Conclusions: The proposed SR-CycleGAN is usable for the SR of a lung clinical CT into μ CT scale, while conventional CycleGAN output images with low qualitative and quantitative values. More lung micro-anatomy information could be observed to aid diagnosis, such as the shape of bronchioles walls.

DeepImageTranslator: A free, user-friendly graphical interface for image translation using deep-learning and its applications in 3D CT image analysis

The advent of deep-learning has set new standards in an array of image translation applications. At present, the use of these methods often requires computer programming experience. Non-commercial programs with graphical interface usually do not allow users to fully customize their deep-learning pipeline. Therefore, our primary objective is to provide a simple graphical interface that allows researchers with no programming experience to easily create, train, and evaluate custom deep-learning models for image translation. We also aimed to test the applicability of our tool in CT image semantic segmentation and noise reduction. DeepImageTranslator was implemented using the Tkinter library, the standard Python interface to the Tk graphical user interface toolkit; backend computations were implemented using data augmentation packages such as Pillow, Numpy, OpenCV, Augmentor, Tensorflow, and Keras libraries. Convolutional neural networks (CNNs) were trained using DeepImageTranslator. The effects of data augmentation, deep-supervision, and sample size on model accuracy were also systematically assessed. The DeepImageTranslator a simple tool that allows users to customize all aspects of their deep-learning pipeline, including the CNN, training optimizer, loss function, and the types of training image augmentation scheme. We showed that DeepImageTranslator can be used to achieve state-of-the-art accuracy and generalizability in semantic segmentation and noise reduction. Highly accurate 3D segmentation models for body composition can be obtained using training sample sizes as small as 17 images. In conclusion, an open-source deep-learning tool for accurate image translation with a user-friendly graphical interface was presented and evaluated. This standalone software can be downloaded at: https://sourceforge.net/projects/deepimagetranslator/.

Structure attention co-training neural network for neovascularization segmentation in intravascular optical coherence tomography

PURPOSE

To development and validate a Neovascularization (NV) segmentation model in intravascular optical coherence tomography (IVOCT) through deep learning methods.

METHODS AND MATERIALS

A total of 1950 2D slices of 70 IVOCT pullbacks were used in our study. We randomly selected 1273 2D slices from 44 patients as the training set, 379 2D slices from 11 patients as the validation set, and 298 2D slices from the last 15 patients as the testing set. Automatic NV segmentation is quite challenging, as it must address issues of speckle noise, shadow artifacts, high distribution variation, etc. To meet these challenges, a new deep learning-based segmentation method is developed based on a co-training architecture with an integrated structural attention mechanism. Co-training is developed to exploit the features of three consecutive slices. The structural attention mechanism comprises spatial and channel attention modules and is integrated into the co-training architecture at each up-sampling step. A cascaded fixed network is further incorporated to achieve segmentation at the image level in a coarse-to-fine manner.

RESULTS

Extensive experiments were performed involving a comparison with several state-of-the-art deep learning-based segmentation methods. Moreover, the consistency of the results with those of manual segmentation was also investigated. Our proposed NV automatic segmentation method achieved the highest correlation with the manual delineation by interventional cardiologists (the Pearson correlation coefficient is 0.825).

CONCLUSION

In this work, we proposed a co-training architecture with an integrated structural attention mechanism to segment NV in IVOCT images. The good agreement between our segmentation results and manual segmentation indicates that the proposed method has great potential for application in the clinical investigation of NV-related plaque diagnosis and treatment. This article is protected by copyright. All rights reserved.