Image Recovery from Synthetic Noise Artifacts in CT Scans Using Modified U-Net.
Ontology highlight
ABSTRACT:
SUBMITTER: Gunawan R
PROVIDER: S-EPMC9505882 | biostudies-literature | 2022 Sep
REPOSITORIES: biostudies-literature
Publications
Image Recovery from Synthetic Noise Artifacts in CT Scans Using Modified U-Net.
Gunawan Rudy R Tran Yvonne Y Zheng Jinchuan J Nguyen Hung H Chai Rifai R
Sensors (Basel, Switzerland) 20220916 18
Computed Tomography (CT) is commonly used for cancer screening as it utilizes low radiation for the scan. One problem with low-dose scans is the noise artifacts associated with low photon count that can lead to a reduced success rate of cancer detection during radiologist assessment. The noise had to be removed to restore detail clarity. We propose a noise removal method using a new model Convolutional Neural Network (CNN). Even though the network training time is long, the result is better than ...[more]
Similar Datasets
Project description:Error estimation in nonlinear medical image registration is a nontrivial problem that is important for validation of registration methods. We propose a supervised method for estimation of registration errors in nonlinear registration of three-dimensional (3-D) images. The method is based on a 3-D convolutional neural network that learns to estimate registration errors from a pair of image patches. By applying the network to patches centered around every voxel, we construct registration error maps. The network is trained using a set of representative images that have been synthetically transformed to construct a set of image pairs with known deformations. The method is evaluated on deformable registrations of inhale-exhale pairs of thoracic CT scans. Using ground truth target registration errors on manually annotated landmarks, we evaluate the method's ability to estimate local registration errors. Estimation of full domain error maps is evaluated using a gold standard approach. The two evaluation approaches show that we can train the network to robustly estimate registration errors in a predetermined range, with subvoxel accuracy. We achieved a root-mean-square deviation of 0.51 mm from gold standard registration errors and of 0.66 mm from ground truth landmark registration errors.
| S-EPMC5943620 | biostudies-other
Project description:BackgroundDaily or weekly cone-beam computed tomography (CBCT) scans are commonly used for accurate patient positioning during the image-guided radiotherapy (IGRT) process, making it an ideal option for adaptive radiotherapy (ART) replanning. However, the presence of severe artifacts and inaccurate Hounsfield unit (HU) values prevent its use for quantitative applications such as organ segmentation and dose calculation. To enable the clinical practice of online ART, it is crucial to obtain CBCT scans with a quality comparable to that of a CT scan.PurposeThis work aims to develop a conditional diffusion model to perform image translation from the CBCT to the CT distribution for the image quality improvement of CBCT.MethodsThe proposed method is a conditional denoising diffusion probabilistic model (DDPM) that utilizes a time-embedded U-net architecture with residual and attention blocks to gradually transform the white Gaussian noise sample to the target CT distribution conditioned on the CBCT. The model was trained on deformed planning CT (dpCT) and CBCT image pairs, and its feasibility was verified in brain patient study and head-and-neck (H&N) patient study. The performance of the proposed algorithm was evaluated using mean absolute error (MAE), peak signal-to-noise ratio (PSNR) and normalized cross-correlation (NCC) metrics on generated synthetic CT (sCT) samples. The proposed method was also compared to four other diffusion model-based sCT generation methods.ResultsIn the brain patient study, the MAE, PSNR, and NCC of the generated sCT were 25.99 HU, 30.49 dB, and 0.99, respectively, compared to 40.63 HU, 27.87 dB, and 0.98 of the CBCT images. In the H&N patient study, the metrics were 32.56 HU, 27.65 dB, 0.98 and 38.99 HU, 27.00, 0.98 for sCT and CBCT, respectively. Compared to the other four diffusion models and one Cycle generative adversarial network (Cycle GAN), the proposed method showed superior results in both visual quality and quantitative analysis.ConclusionsThe proposed conditional DDPM method can generate sCT from CBCT with accurate HU numbers and reduced artifacts, enabling accurate CBCT-based organ segmentation and dose calculation for online ART.
| S-EPMC11747767 | biostudies-literature
Project description:Image denoising is a complex task that always yields an approximated version of the clean image. Unfortunately, the existing works have focussed only on the peak signal to noise ratio (PSNR) metric and have shown no attention to edge features in a reconstructed image. Although fully convolution neural networks (CNN) are capable of removing the noise using kernel filters and automatic extraction of features, it has failed to reconstruct the images for higher values of noise standard deviation. Additionally, deep learning models require a huge database to learn better from the inputs. This, in turn, increases the computational complexity and memory requirement. Therefore, we propose the Median Noise Residue U-Net (MNRU-Net) with a limited training database without involving image augmentation. In the proposed work, the learning capability of the traditional U-Net model was increased by adding hand-crafted features in the input layers of the U-Net. Further, an approximate version of the noise estimated from the median filter and the gradient information of the image were used to improve the performance of U-Net. Later, the performance of MNRU-Net was evaluated based on PSNR, structural similarity, and figure of merit for different noise standard deviations of 15, 25, and 50 respectively. It is witnessed that the results gained from the suggested work are better than the results yielded by complex denoising models such as the robust deformed denoising CNN (RDDCNN). This work emphasizes that the skip connections along with the hand-crafted features could improve the performance at higher noise levels by using this simple architecture. In addition, the model was found to be less expensive, with low computational complexity.
| S-EPMC11784896 | biostudies-literature
Project description:Purpose: To quantify the impact of image noise on CT-based lung ventilation biomarkers calculated using Jacobian determinant techniques. Methods: Five mechanically ventilated swine were imaged on a multi-row CT scanner with acquisition parameters of 120 kVp and 0.6 mm slice thickness in static and 4-dimensional CT (4DCT) modes with respective pitches of 1 and 0.09. A range of tube current time product (mAs) values were used to vary image dose. On two dates, subjects received two 4DCTs: one with 10 mAs/rotation (low-dose, high-noise) and one with CT simulation standard of care 100 mAs/rotation (high-dose, low-noise). Additionally, 10 intermediate noise level breath-hold (BHCT) scans were acquired with inspiratory and expiratory lung volumes. Images were reconstructed with and without iterative reconstruction (IR) using 1 mm slice thickness. The Jacobian determinant of an estimated transformation from a B-spline deformable image registration was used to create CT-ventilation biomarkers estimating lung tissue expansion. 24 CT-ventilation maps were generated per subject per scan date: four 4DCT ventilation maps (two noise levels each with and without IR) and 20 BHCT ventilation maps (10 noise levels each with and without IR). Biomarkers derived from reduced dose scans were registered to the reference full dose scan for comparison. Evaluation metrics were gamma pass rate (Γ) with 2 mm distance-to-agreement and 6% intensity criterion, voxel-wise Spearman correlation (ρ) and Jacobian ratio coefficient of variation (CoVJR). Results: Comparing biomarkers derived from low (CTDIvol = 6.07 mGy) and high (CTDIvol = 60.7 mGy) dose 4DCT scans, mean Γ, ρ and CoVJR values were 93% ± 3%, 0.88 ± 0.03 and 0.04 ± 0.009, respectively. With IR applied, those values were 93% ± 4%, 0.90 ± 0.04 and 0.03 ± 0.003. Similarly, comparisons between BHCT-based biomarkers with variable dose (CTDIvol = 1.35–7.95 mGy) had mean Γ, ρ and CoVJR of 93% ± 4%, 0.97 ± 0.02 and 0.03 ± 0.006 without IR and 93% ± 4%, 0.97 ± 0.03 and 0.03 ± 0.007 with IR. Applying IR did not significantly change any metrics (p Discussion: This work demonstrated that CT-ventilation, calculated using the Jacobian determinant of an estimated transformation from a B-spline deformable image registration, is invariant to Hounsfield Unit (HU) variation caused by image noise. This advantageous finding may be leveraged clinically with potential applications including dose reduction and/or acquiring repeated low-dose acquisitions for improved ventilation characterization.
| S-EPMC9971492 | biostudies-literature
Project description:Purpose/objectiveAlthough radiotherapy is a key component of cancer treatment, its implementation into pre-clinical in vivo models with relatively small target volumes is frequently omitted either due to technical complexity or expected side effects hampering long-term observational studies. We here demonstrate how an affordable industrial micro-CT can be converted into a small animal IGRT device at very low costs. We also demonstrate the proof of principle for the case of partial brain irradiation of mice carrying orthotopic glioblastoma implants.Methods/materialsA commercially available micro-CT originally designed for non-destructive material analysis was used. It consists of a CNC manipulator, a transmission X-ray tube (10-160 kV) and a flat-panel detector, which was used together with custom-made steel collimators (1-5 mm aperture size). For radiation field characterization, an ionization chamber, water-equivalent slab phantoms and radiochromic films were used. A treatment planning tool was implemented using a C++ application. For proof of principle, NOD/SCID/γc(-/-) mice were orthotopically implanted with U87MG high-grade glioma cells and irradiated using the novel setup.ResultsThe overall symmetry of the radiation field at 150 kV was 1.04 ± 0.02%. The flatness was 4.99 ± 0.63% and the penumbra widths were between 0.14 mm and 0.51 mm. The full width at half maximum (FWHM) ranged from 1.97 to 9.99 mm depending on the collimator aperture size. The dose depth curve along the central axis followed a typical shape of keV photons. Dose rates measured were 10.7 mGy/s in 1 mm and 7.6 mGy/s in 5 mm depth (5 mm collimator aperture size). Treatment of mice with a single dose of 10 Gy was tolerated well and resulted in central tumor necrosis consistent with therapeutic efficacy.ConclusionA conventional industrial micro-CT can be easily modified to allow effective small animal IGRT even of critical target volumes such as the brain.
| S-EPMC4438006 | biostudies-literature
Project description:BackgroundCone-beam computed tomography (CBCT) imaging of the brain can be performed in the angiography suite to support various neurovascular procedures. Relying on CBCT brain imaging solely, however, still lacks full diagnostic confidence due to the inferior image quality compared with CT and various imaging artifacts that persist even with modern CBCT.ObjectiveTo perform a detailed evaluation of image artifact improvement using a new CBCT protocol which implements a novel dual-axis 'butterfly' trajectory.MethodsOur study included 94 scans from 47 patients who received CBCT imaging for assessment of either ischemia or hemorrhage during a neurovascular procedure. Both a traditional uni-axis 'circular' and novel dual-axis 'butterfly' protocol were performed on each patient (same-patient control). Each brain scan was divided into six regions and scored out of 3 based on six artifacts originating from various physics-based and patient-based sources.ResultsThe dual-axis trajectory produces CBCT images with significantly fewer image artifacts than the traditional circular scan (whole brain average artifact score, AS: 0.20 vs 0.33), with the greatest improvement in bone beam hardening (AS: 0.13 vs 0.78) and cone-beam artifacts (AS: 0.04 vs 0.55).ConclusionsRecent developments in CBCT imaging protocols have significantly improved image artifacts, which has improved diagnostic confidence for stroke and supports a direct-to-angiography suite transfer approach for patients with acute ischemic stroke.
| S-EPMC9985729 | biostudies-literature
Project description:PurposeTo enhance the image quality of oncology [18F]-FDG PET scans acquired in shorter times and reconstructed by faster algorithms using deep neural networks.MethodsList-mode data from 277 [18F]-FDG PET/CT scans, from six centres using GE Discovery PET/CT scanners, were split into ¾-, ½- and ¼-duration scans. Full-duration datasets were reconstructed using the convergent block sequential regularised expectation maximisation (BSREM) algorithm. Short-duration datasets were reconstructed with the faster OSEM algorithm. The 277 examinations were divided into training (n = 237), validation (n = 15) and testing (n = 25) sets. Three deep learning enhancement (DLE) models were trained to map full and partial-duration OSEM images into their target full-duration BSREM images. In addition to standardised uptake value (SUV) evaluations in lesions, liver and lungs, two experienced radiologists scored the quality of testing set images and BSREM in a blinded clinical reading (175 series).ResultsOSEM reconstructions demonstrated up to 22% difference in lesion SUVmax, for different scan durations, compared to full-duration BSREM. Application of the DLE models reduced this difference significantly for full-, ¾- and ½-duration scans, while simultaneously reducing the noise in the liver. The clinical reading showed that the standard DLE model with full- or ¾-duration scans provided an image quality substantially comparable to full-duration scans with BSREM reconstruction, yet in a shorter reconstruction time.ConclusionDeep learning-based image enhancement models may allow a reduction in scan time (or injected activity) by up to 50%, and can decrease reconstruction time to a third, while maintaining image quality.
| S-EPMC8803788 | biostudies-literature
Project description:Feedback from radiologists indicated that differences in image appearance and noise impeded reading of post-contrast computed tomography (CT) scans from an updated CT scanner that was recently added to a fleet of existing scanners from the same vendor, despite using identically named reconstruction algorithms. The goals of this work were to quantify and possibly standardize image quality on the new and an existing scanner using phantom images. Three months of daily quality control images were analyzed to determine the mean CT number and noise magnitude in a water phantom. Next, subtraction images from the uniformity section of an American College of Radiology CT phantom were used to generate noise power spectra for both scanners. Then, a semi-anthropomorphic liver phantom was imaged with both scanners in triplicate using identical body protocols to quantify differences CT number and noise magnitude. Finally, the scanner dependence of CT number and noise magnitude on material attenuation was quantified using a multi-energy CT phantom with 15 material inserts. Significant differences between scanners were determined using a paired or Welch's t test as appropriate. In daily quality control images, the new scanner exhibited slightly higher CT number (0.697 vs. 0.412, P < 0.001, n = 85) and slightly lower noise magnitude (4.85 vs. 4.94, P < 0.001, n = 85). Measured NPS was not significantly different between the existing and new scanners. Interestingly, it was observed that the noise magnitude from the new scanner increased with increasing material attenuation in both the liver (P = 0.008) and multi-energy (P < 0.001) phantoms. Using an alternate reconstruction algorithm with the new scanner eliminated this deviation at high material attenuations. While standard noise evaluation in a water phantom was unable to discern differences between the scanners, more comprehensive testing with higher attenuation materials allowed for the characterization and homogenization of image quality.
| S-EPMC6964752 | biostudies-literature
Project description:Background and purposeThere is a desire within many institutions to reduce the radiation dose in CTP examinations. The purpose of this study was to simulate dose reduction through the addition of noise in brain CT perfusion examinations and to determine the subsequent effects on quality and quantitative interpretation.Materials and methodsA total of 22 consecutive reference CTP scans were identified from an institutional review board-approved prospective clinical trial, all performed at 80 keV and 190 mAs. Lower-dose scans at 188, 177, 167, 127, and 44 mAs were generated through the addition of spatially correlated noise to the reference scans. A standard software package was used to generate CBF, CBV, and MTT maps. Six blinded radiologists determined quality scores of simulated scans on a Likert scale. Quantitative differences were calculated.ResultsFor qualitative analysis, the correlation coefficients for CBF (-0.34; P < .0001), CBV (-0.35; P < .0001), and MTT (-0.44; P < .0001) were statistically significant. Interobserver agreements in quality for the simulated 188-, 177-, 167-, 127-, and 44-mAs scans for CBF were 0.95, 0.98, 0.98, 0.95, and 0.52, respectively. Interobserver agreements in quality for the simulated CBV were 1, 1, 1, 1, and 0.83, respectively. For MTT, the interobserver agreements were 0.83, 0.86, 0.88, 0.74, and 0.05, respectively. For quantitative analysis, only the lowest simulated dose of 44 mAs showed statistically significant differences from the reference scan values for CBF (-1.8; P = .04), CBV (0.07; P < .0001), and MTT (0.46; P < .0001).ConclusionsFrom a reference CTP study performed at 80 keV and 190 mAs, this simulation study demonstrates the potential of a 33% reduction in tube current and dose while maintaining image quality and quantitative interpretations. This work can be used to inform future studies by using true, nonsimulated scans.
| S-EPMC4108445 | biostudies-literature
Project description:ObjectiveTo assess the feasibility of applying a deep learning-based denoising technique to coronary CT angiography (CCTA) along with iterative reconstruction for additional noise reduction.Materials and methodsWe retrospectively enrolled 82 consecutive patients (male:female = 60:22; mean age, 67.0 ± 10.8 years) who had undergone both CCTA and invasive coronary artery angiography from March 2017 to June 2018. All included patients underwent CCTA with iterative reconstruction (ADMIRE level 3, Siemens Healthineers). We developed a deep learning based denoising technique (ClariCT.AI, ClariPI), which was based on a modified U-net type convolutional neural net model designed to predict the possible occurrence of low-dose noise in the originals. Denoised images were obtained by subtracting the predicted noise from the originals. Image noise, CT attenuation, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were objectively calculated. The edge rise distance (ERD) was measured as an indicator of image sharpness. Two blinded readers subjectively graded the image quality using a 5-point scale. Diagnostic performance of the CCTA was evaluated based on the presence or absence of significant stenosis (≥ 50% lumen reduction).ResultsObjective image qualities (original vs. denoised: image noise, 67.22 ± 25.74 vs. 52.64 ± 27.40; SNR [left main], 21.91 ± 6.38 vs. 30.35 ± 10.46; CNR [left main], 23.24 ± 6.52 vs. 31.93 ± 10.72; all p < 0.001) and subjective image quality (2.45 ± 0.62 vs. 3.65 ± 0.60, p < 0.001) improved significantly in the denoised images. The average ERDs of the denoised images were significantly smaller than those of originals (0.98 ± 0.08 vs. 0.09 ± 0.08, p < 0.001). With regard to diagnostic accuracy, no significant differences were observed among paired comparisons.ConclusionApplication of the deep learning technique along with iterative reconstruction can enhance the noise reduction performance with a significant improvement in objective and subjective image qualities of CCTA images.
| S-EPMC7458859 | biostudies-literature