Routine-dose Images Research Articles

Ultra-low-dose CT lung screening with artificial intelligence iterative reconstruction: evaluation via automatic nodule-detection software

To test the feasibility of ultra-low-dose (ULD) computed tomography (CT) combined with an artificial intelligence iterative reconstruction (AIIR) algorithm for screening pulmonary nodules using computer-assisted diagnosis (CAD). A chest phantom with artificial pulmonary nodules was first scanned using the routine protocol and the ULD protocol (3.28 versus 0.18 mSv) to compare the image quality and to test the acceptability of the ULD CT protocol. Next, 147 lung-screening patients were enrolled prospectively, undergoing an additional ULD CT immediately after their routine CT examination for clinical validation. Images were reconstructed with filtered back-projection (FBP), hybrid iterative reconstruction (HIR), the AIIR, and were imported to the CAD software for preliminary nodule detection. Subjective image quality on the phantom was scored using a five-point scale and compared using the Mann-Whitney U-test. Nodule detection using CAD was evaluated for ULD HIR and AIIR images using the routine dose image as reference. Higher image quality was scored for AIIR than for FBP and HIR at ULD (p<0.001). As reported by CAD, 107 patients were presented with more than five nodules on routine dose images and were chosen to represent the challenging cases at an early stage of pulmonary disease. Among such, the performance of nodule detection by CAD on ULD HIR and AIIR images was 75.2% and 92.2% of the routine dose image, respectively. Combined with AIIR, it was feasible to use an ULD CT protocol with 95% dose reduction for CAD-based screening of pulmonary nodules.

Experimental and clinical validation of an artificial intelligence metal artifact correction algorithm for low-dose following up CT of percutaneous vertebroplasty.

Low-dose following up computed tomography (CT) of percutaneous vertebroplasty (PVP) that involves the use of bone cement usually suffers from lightweight metal artifacts, where conventional techniques for CT metal artifact reduction are often not sufficiently effective. This study aimed to validate an artificial intelligence (AI)-based metal artifact correction (MAC) algorithm for use in low-dose following up CT for PVP. In experimental validation, an ovine vertebra phantom was designed to simulate the clinical scenario of PVP. With routine-dose images acquired prior to the cement introduction as the reference, low-dose CT scans were taken on the cemented phantom and processed with conventional MAC and AI-MAC. The resulting image quality was compared in CT attenuation, image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR), followed by a quantitative evaluation of the artifact correction accuracy based on adaptive segmentation of the paraspinal muscle. In clinical validation, ten cases of low-dose following up CT after PVP were enrolled to test the performance of diagnosing sarcopenia with measured CT attenuation per cemented vertebral segment, via receiver operating characteristic (ROC) analysis. With respect to the reference image, no significant difference was found for AI-MAC in CT attenuation, image noise, SNRs, and CNR (all P>0.05). The paraspinal muscle segmented on the AI-MAC image was 18.6% and 8.3% more complete to uncorrected and MAC images. Higher area under the curve (AUC) of the ROC analysis was found for AI-MAC (AUC =0.92) compared to the uncorrected (AUC =0.61) and MAC images (AUC =0.70). In low-dose following up CT for PVP, the AI-MAC has been fully validated for its superior ability compared to conventional MAC in suppressing artifacts and may be a reliable alternative for diagnosing sarcopenia.

Diagnostic Performance in Low- and High-Contrast Tasks of an Image-Based Denoising Algorithm Applied to Radiation Dose-Reduced Multiphase Abdominal CT Examinations.

BACKGROUND. Anatomic redundancy between phases can be used to achieve denoising of multiphase CT examinations. A limitation of iterative reconstruction (IR) techniques is that they generally require use of CT projection data. A frequency-split multi-band-filtration algorithm applies denoising to the multiphase CT images themselves. This method does not require knowledge of the acquisition process or integration into the reconstruction system of the scanner, and it can be implemented as a supplement to commercially available IR algorithms. OBJECTIVE. The purpose of the present study is to compare radiologists' performance for low-contrast and high-contrast diagnostic tasks (i.e., tasks for which differences in CT attenuation between the imaging target and its anatomic background are subtle or large, respectively) evaluated on multiphase abdominal CT between routine-dose images and radiation dose-reduced images processed by a frequency-split multiband-filtration denoising algorithm. METHODS. This retrospective single-center study included 47 patients who underwent multiphase contrast-enhanced CT for known or suspected liver metastases (a low-contrast task) and 45 patients who underwent multiphase contrast-enhanced CT for pancreatic cancer staging (a high-contrast task). Radiation dose-reduced images corresponding to dose reduction of 50% or more were created using a validated noise insertion technique and then underwent denoising using the frequency-split multi-band-filtration algorithm. Images were independently evaluated in multiple sessions by different groups of abdominal radiologists for each task (three readers in the low-contrast arm and four readers in the high-contrast arm). The noninferiority of denoised radiation dose-reduced images to routine-dose images was assessed using the jackknife alternative free-response ROC (JAFROC) figure-of-merit (FOM; limit of noninferiority, -0.10) for liver metastases detection and using the Cohen kappa statistic and reader confidence scores (100-point scale) for pancreatic cancer vascular invasion. RESULTS. For liver metastases detection, the JAFROC FOM for denoised radiation dose-reduced images was 0.644 (95% CI, 0.510-0.778), and that for routine-dose images was 0.668 (95% CI, 0.543-0.792; estimated difference, -0.024 [95% CI, -0.084 to 0.037]). Intraobserver agreement for pancreatic cancer vascular invasion was substantial to near perfect when the two image sets were compared (κ = 0.53-1.00); the 95% CIs of all differences in confidence scores between image sets contained zero. CONCLUSION. Multiphase contrast-enhanced abdominal CT images with a radiation dose reduction of 50% or greater that undergo denoising by a frequency-split multiband-filtration algorithm yield performance similar to that of routine-dose images for detection of liver metastases and vascular staging of pancreatic cancer. CLINICAL IMPACT. The image-based denoising algorithm facilitates radiation dose reduction of multiphase examinations for both low- and high-contrast diagnostic tasks without requiring manufacturer-specific hardware or software.

Prior iterative reconstruction (PIR) to lower radiation dose and preserve radiologist performance for multiphase liver CT: a multi-reader pilot study.

Prior iterative reconstruction (PIR) spatially registers CT image data from multiple phases of enhancement to reduce image noise. We evaluated PIR in contrast-enhanced multiphase liver CT. Patients with archived projection CT data with proven malignant or benign liver lesions, or without lesions, by reference criteria were included. Lower-dose PIR images were reconstructed using validated noise insertion from multiphase CT exams (50% dose in 2 phases, 25% dose in 1 phase). The phase of enhancement most relevant to the diagnostic task was selected for evaluation. Four radiologists reviewed routine-dose and lower-dose PIR images, circumscribing liver lesions and rating confidence for malignancy (0 to 100) and image quality. JAFROC Figures of Merit (FOM) were calculated. 31 patients had 60 liver lesions (28 primary hepaticmalignancies, 6 hepatic metastases, 26 benign lesions). Pooled JAFROC FOM for malignancy for routine-dose CT was 0.615 (95% CI 0.464, 0.767) compared to 0.662 for PIR (95% CI 0.527, 0.797). The estimated FOM difference between the routine-dose and lower-dose PIR images was + 0.047 (95% CI - 0.023, + 0.116). Pooled sensitivity/specificity for routine-dose images was 70%/68% compared to 73%/66% for lower-dose PIR. Lower-dose PIR had lower diagnostic image quality (mean 3.8 vs. 4.2, p = 0.0009) and sharpness (mean 2.3 vs. 2.0, p = 0.0071). PIR is a promising method to reduce radiation dose for multiphase abdominal CT, preserving observer performance despite small reductions in image quality. Further work is warranted.

Generative Adversarial Networks for Noise Reduction in Low-Dose CT.

Noise is inherent to low-dose CT acquisition. We propose to train a convolutional neural network (CNN) jointly with an adversarial CNN to estimate routine-dose CT images from low-dose CT images and hence reduce noise. A generator CNN was trained to transform low-dose CT images into routine-dose CT images using voxelwise loss minimization. An adversarial discriminator CNN was simultaneously trained to distinguish the output of the generator from routine-dose CT images. The performance of this discriminator was used as an adversarial loss for the generator. Experiments were performed using CT images of an anthropomorphic phantom containing calcium inserts, as well as patient non-contrast-enhanced cardiac CT images. The phantom and patients were scanned at 20% and 100% routine clinical dose. Three training strategies were compared: the first used only voxelwise loss, the second combined voxelwise loss and adversarial loss, and the third used only adversarial loss. The results showed that training with only voxelwise loss resulted in the highest peak signal-to-noise ratio with respect to reference routine-dose images. However, CNNs trained with adversarial loss captured image statistics of routine-dose images better. Noise reduction improved quantification of low-density calcified inserts in phantom CT images and allowed coronary calcium scoring in low-dose patient CT images with high noise levels. Testing took less than 10 s per CT volume. CNN-based low-dose CT noise reduction in the image domain is feasible. Training with an adversarial network improves the CNNs ability to generate images with an appearance similar to that of reference routine-dose CT images.

Observer Performance in the Detection and Classification of Malignant Hepatic Nodules and Masses with CT Image-Space Denoising and Iterative Reconstruction.

To determine if lower-dose computed tomographic (CT) scans obtained with adaptive image-based noise reduction (adaptive nonlocal means [ANLM]) or iterative reconstruction (sinogram-affirmed iterative reconstruction [SAFIRE]) result in reduced observer performance in the detection of malignant hepatic nodules and masses compared with routine-dose scans obtained with filtered back projection (FBP). This study was approved by the institutional review board and was compliant with HIPAA. Informed consent was obtained from patients for the retrospective use of medical records for research purposes. CT projection data from 33 abdominal and 27 liver or pancreas CT examinations were collected (median volume CT dose index, 13.8 and 24.0 mGy, respectively). Hepatic malignancy was defined by progression or regression or with histopathologic findings. Lower-dose data were created by using a validated noise insertion method (10.4 mGy for abdominal CT and 14.6 mGy for liver or pancreas CT) and images reconstructed with FBP, ANLM, and SAFIRE. Four readers evaluated routine-dose FBP images and all lower-dose images, circumscribing liver lesions and selecting diagnosis. The jackknife free-response receiver operating characteristic figure of merit (FOM) was calculated on a per-malignant nodule or per-mass basis. Noninferiority was defined by the lower limit of the 95% confidence interval (CI) of the difference between lower-dose and routine-dose FOMs being less than -0.10. Twenty-nine patients had 62 malignant hepatic nodules and masses. Estimated FOM differences between lower-dose FBP and lower-dose ANLM versus routine-dose FBP were noninferior (difference: -0.041 [95% CI: -0.090, 0.009] and -0.003 [95% CI: -0.052, 0.047], respectively). In patients with dedicated liver scans, lower-dose ANLM images were noninferior (difference: +0.015 [95% CI: -0.077, 0.106]), whereas lower-dose FBP images were not (difference -0.049 [95% CI: -0.140, 0.043]). In 37 patients with SAFIRE reconstructions, the three lower-dose alternatives were found to be noninferior to the routine-dose FBP. At moderate levels of dose reduction, lower-dose FBP images without ANLM or SAFIRE were noninferior to routine-dose images for abdominal CT but not for liver or pancreas CT.

Pilot Study of Detection, Radiologist Confidence and Image Quality With Sinogram-Affirmed Iterative Reconstruction at Half–Routine Dose Level

The objective of this study was to determine the effect of Sinogram-Affirmed Iterative Reconstruction (SAFIRE) on radiological detection, diagnostic confidence, and image quality at half-dose, contrast-enhanced abdominopelvic computed tomography. Forty dual-source examinations were reconstructed using routine-dose with filtered back projection, half-dose filtered back projection, and half-dose SAFIRE. Three radiologists detected lesions in abdominopelvic organs, reporting findings of potential medical significance, diagnostic confidence, and image quality. There was greater than 78% concordance between full- and half-dose images ± SAFIRE, and no difference in the detection of lesions within organs between half-dose images ± SAFIRE (P = 0.22 - 1.0). Detection of potentially important findings varied by reader, but not between dose/reconstruction methods. Diagnostic confidence varied widely (P < 0.001 to P > 0.91). Sinogram-Affirmed Iterative Reconstruction significantly improved image quality in the pelvis (P ≤ 0.04). Half-dose images ± SAFIRE had organ-specific detections similar to routine-dose images. Sinogram-Affirmed Iterative Reconstruction improved image quality in the pelvis, but diagnostic confidence and image quality scores in the abdomen depended on the reader.