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

Development and Validation of an Artificial Intelligence Model for Detecting Rib Fractures on Chest Radiographs

Background: Chest radiography is the standard method for detecting rib fractures. Our study aims to develop an artificial intelligence (AI) model that, with only a relatively small amount of training data, can identify rib fractures on chest radiographs and accurately mark their precise locations, thereby achieving a diagnostic accuracy comparable to that of medical professionals. Methods: For this retrospective study, we developed an AI model using 540 chest radiographs (270 normal and 270 with rib fractures) labeled for use with Detectron2 which incorporates a faster region-based convolutional neural network (R-CNN) enhanced with a feature pyramid network (FPN). The model's ability to classify radiographs and detect rib fractures was assessed. Furthermore, we compared the model's performance to that of 12 physicians, including six board-certified anesthesiologists and six residents, through an observer performance test. Results: Regarding the radiographic classification performance of the AI model, the sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were 0.87, 0.83, and 0.89, respectively. In terms of rib fracture detection performance, the sensitivity, false-positive rate, and free-response receiver operating characteristic (JAFROC) figure of merit (FOM) were 0.62, 0.3, and 0.76, respectively. The AI model showed no statistically significant difference in the observer performance test compared to 11 of 12 and 10 of 12 physicians, respectively. Conclusions: We developed an AI model trained on a limited dataset that demonstrated a rib fracture classification and detection performance comparable to that of an experienced physician.

Image Quality and Diagnostic Performance of Low-Dose Liver CT with Deep Learning Reconstruction versus Standard-Dose CT

Purpose To compare the image quality and diagnostic capability in detecting malignant liver tumors of low-dose CT (LDCT, 33% dose) with deep learning-based denoising (DLD) and standard-dose CT (SDCT, 100% dose) with model-based iterative reconstruction (MBIR). Materials and Methods In this prospective, multicenter, noninferiority study, individuals referred for liver CT scans were enrolled from three tertiary referral hospitals between February 2021 and August 2022. All liver CT scans were conducted using a dual-source scanner with the dose split into tubes A (67% dose) and B (33% dose). Blended images from tubes A and B were created using MBIR to produce SDCT images, whereas LDCT images used data from tube B and were reconstructed with DLD. The noise in liver images was measured and compared between imaging techniques. The diagnostic performance of each technique in detecting malignant liver tumors was evaluated by three independent radiologists using jackknife alternative free-response receiver operating characteristic analysis. Noninferiority of LDCT compared with SDCT was declared when the lower limit of the 95% CI for the difference in figure of merit (FOM) was greater than -0.10. Results A total of 296 participants (196 men, 100 women; mean age, 60.5 years ± 13.3 [SD]) were included. The mean noise level in the liver was significantly lower for LDCT (10.1) compared with SDCT (10.7) (P < .001). Diagnostic performance was assessed in 246 participants (108 malignant tumors in 90 participants). The reader-averaged FOM was 0.880 for SDCT and 0.875 for LDCT (P = .35). The difference fell within the noninferiority margin (difference, -0.005 [95% CI: -0.024, 0.012]). Conclusion Compared with SDCT with MBIR, LDCT using 33% of the standard radiation dose had reduced image noise and comparable diagnostic performance in detecting malignant liver tumors. Keywords: CT, Abdomen/GI, Liver, Comparative Studies, Diagnosis, Reconstruction Algorithms Clinical trial registration no. NCT05804799 © RSNA, 2024 Supplemental material is available for this article.

Milestones in CT: Past, Present, and Future

In 1971, the first patient CT examination by Ambrose and Hounsfield paved the way for not only volumetric imaging of the brain but of the entire body. From the initial 5-minute scan for a 180° rotation to today's 0.24-second scan for a 360° rotation, CT technology continues to reinvent itself. This article describes key historical milestones in CT technology from the earliest days of CT to the present, with a look toward the future of this essential imaging modality. After a review of the beginnings of CT and its early adoption, the technical steps taken to decrease scan times-both per image and per examination-are reviewed. Novel geometries such as electron-beam CT and dual-source CT have also been developed in the quest for ever-faster scans and better in-plane temporal resolution. The focus of the past 2 decades on radiation dose optimization and management led to changes in how exposure parameters such as tube current and tube potential are prescribed such that today, examinations are more customized to the specific patient and diagnostic task than ever before. In the mid-2000s, CT expanded its reach from gray-scale to color with the clinical introduction of dual-energy CT. Today's most recent technical innovation-photon-counting CT-offers greater capabilities in multienergy CT as well as spatial resolution as good as 125 μm. Finally, artificial intelligence is poised to impact both the creation and processing of CT images, as well as automating many tasks to provide greater accuracy and reproducibility in quantitative applications.

75% radiation dose reduction using deep learning reconstruction on low-dose chest CT

OBJECTIVE

Few studies have explored the clinical feasibility of using deep-learning reconstruction to reduce the radiation dose of CT. We aimed to compare the image quality and lung nodule detectability between chest CT using a quarter of the low dose (QLD) reconstructed with vendor-agnostic deep-learning image reconstruction (DLIR) and conventional low-dose (LD) CT reconstructed with iterative reconstruction (IR).

MATERIALS AND METHODS

We retrospectively collected 100 patients (median age, 61 years [IQR, 53-70 years]) who received LDCT using a dual-source scanner, where total radiation was split into a 1:3 ratio. QLD CT was generated using a quarter dose and reconstructed with DLIR (QLD-DLIR), while LDCT images were generated using a full dose and reconstructed with IR (LD-IR). Three thoracic radiologists reviewed subjective noise, spatial resolution, and overall image quality, and image noise was measured in five areas. The radiologists were also asked to detect all Lung-RADS category 3 or 4 nodules, and their performance was evaluated using area under the jackknife free-response receiver operating characteristic curve (AUFROC).

RESULTS

The median effective dose was 0.16 (IQR, 0.14-0.18) mSv for QLD CT and 0.65 (IQR, 0.57-0.71) mSv for LDCT. The radiologists' evaluations showed no significant differences in subjective noise (QLD-DLIR vs. LD-IR, lung-window setting; 3.23 ± 0.19 vs. 3.27 ± 0.22; P = .11), spatial resolution (3.14 ± 0.28 vs. 3.16 ± 0.27; P = .12), and overall image quality (3.14 ± 0.21 vs. 3.17 ± 0.17; P = .15). QLD-DLIR demonstrated lower measured noise than LD-IR in most areas (P < .001 for all). No significant difference was found between QLD-DLIR and LD-IR for the sensitivity (76.4% vs. 72.2%; P = .35) or the AUFROCs (0.77 vs. 0.78; P = .68) in detecting Lung-RADS category 3 or 4 nodules. Under a noninferiority limit of -0.1, QLD-DLIR showed noninferior detection performance (95% CI for AUFROC difference, -0.04 to 0.06).

CONCLUSION

QLD-DLIR images showed comparable image quality and noninferior nodule detectability relative to LD-IR images.

Deep learning based automatic detection algorithm for acute intracranial haemorrhage: a pivotal randomized clinical trial

Acute intracranial haemorrhage (AIH) is a potentially life-threatening emergency that requires prompt and accurate assessment and management. This study aims to develop and validate an artificial intelligence (AI) algorithm for diagnosing AIH using brain-computed tomography (CT) images. A retrospective, multi-reader, pivotal, crossover, randomised study was performed to validate the performance of an AI algorithm was trained using 104,666 slices from 3010 patients. Brain CT images (12,663 slices from 296 patients) were evaluated by nine reviewers belonging to one of the three subgroups (non-radiologist physicians, n = 3; board-certified radiologists, n = 3; and neuroradiologists, n = 3) with and without the aid of our AI algorithm. Sensitivity, specificity, and accuracy were compared between AI-unassisted and AI-assisted interpretations using the chi-square test. Brain CT interpretation with AI assistance results in significantly higher diagnostic accuracy than that without AI assistance (0.9703 vs. 0.9471, p < 0.0001, patient-wise). Among the three subgroups of reviewers, non-radiologist physicians demonstrate the greatest improvement in diagnostic accuracy for brain CT interpretation with AI assistance compared to that without AI assistance. For board-certified radiologists, the diagnostic accuracy for brain CT interpretation is significantly higher with AI assistance than without AI assistance. For neuroradiologists, although brain CT interpretation with AI assistance results in a trend for higher diagnostic accuracy compared to that without AI assistance, the difference does not reach statistical significance. For the detection of AIH, brain CT interpretation with AI assistance results in better diagnostic performance than that without AI assistance, with the most significant improvement observed for non-radiologist physicians.

Ultra-low-dose hepatic multiphase CT using deep learning-based image reconstruction algorithm focused on arterial phase in chronic liver disease: A non-inferiority study

PURPOSE

This study determined whether image quality and detectability of ultralow-dose hepatic multiphase CT (ULDCT, 33.3% dose) using a vendor-agnostic deep learning model(DLM) are noninferior to those of standard-dose CT (SDCT, 100% dose) using model-based iterative reconstruction(MBIR) in patients with chronic liver disease focusing on arterial phase.

METHODS

Sixty-seven patients underwent hepatic multiphase CT using a dual-source scanner to obtain two different radiation dose CT scans (100%, SDCT and 33.3%, ULDCT). ULDCT using DLM and SDCT using MBIR were compared. A margin of -0.5 for the difference between the two protocols was pre-defined as noninferiority of the overall image quality of the arterial phase image. Quantitative image analysis (signal to noise ratio[SNR] and contrast to noise ratio[CNR]) was also conducted. The detectability of hepatic arterial focal lesions was compared using the Jackknife free-response receiver operating characteristic analysis. Non-inferiority was satisfied if the margin of the lower limit of 95%CI of the difference in figure-of-merit was less than -0.1.

RESULTS

Mean overall arterial phase image quality scores with ULDCT using DLM and SDCT using MBIR were 4.35 ± 0.57 and 4.08 ± 0.58, showing noninferiority (difference: -0.269; 95 %CI, -0.374 to -0.164). ULDCT using DLM showed a significantly superior contrast-to-noise ratio of arterial enhancing lesion (p < 0.05). Figure-of-merit for detectability of arterial hepatic focal lesion was 0.986 for ULDCT using DLM and 0.963 for SDCT using MBIR, showing noninferiority (difference: -0.023, 95 %CI: -0.016 to 0.063).

CONCLUSION

ULDCT using DLM with 66.7% dose reduction showed non-inferior overall image quality and detectability of arterial focal hepatic lesion compared to SDCT using MBIR.

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.

Oncology-specific radiation dose and image noise reference levels in adult abdominal-pelvic CT

OBJECTIVES

To provide our oncology-specific adult abdominal-pelvic CT reference levels for image noise and radiation dose from a high-volume, oncologic, tertiary referral center.

METHODS

The portal venous phase abdomen-pelvis acquisition was assessed for image noise and radiation dose in 13,320 contrast-enhanced CT examinations. Patient size (effective diameter) and radiation dose (CTDIvol) were recorded using a commercial software system, and image noise (Global Noise metric) was quantified using a custom processing system. The reference level and range for dose and noise were calculated for the full dataset, and for examinations grouped by CT scanner model. Dose and noise reference levels were also calculated for exams grouped by five different patient size categories.

RESULTS

The noise reference level was 11.25 HU with a reference range of 10.25-12.25 HU. The dose reference level at a median effective diameter of 30.7 cm was 26.7 mGy with a reference range of 19.6-37.0 mGy. Dose increased with patient size; however, image noise remained approximately constant within the noise reference range. The doses were 2.1-2.5 times than the doses in the ACR DIR registry for corresponding patient sizes. The image noise was 0.63-0.75 times the previously published reference level in abdominal-pelvic CT examinations.

CONCLUSIONS

Our oncology-specific abdominal-pelvic CT dose reference levels are higher than in the ACR dose index registry and our oncology-specific image noise reference levels are lower than previously proposed image noise reference levels.

ADVANCES IN KNOWLEDGE

This study reports reference image noise and radiation dose levels appropriate for the indication of abdomen-pelvis CT examination for cancer diagnosis and staging. The difference in these reference levels from non-oncology-specific CT examinations highlight a need for indication-specific, dose index and image quality reference registries.

A Review of Deep Learning CT Reconstruction: Concepts, Limitations, and Promise in Clinical Practice

Purpose of Review

Deep Learning reconstruction (DLR) is the current state-of-the-art method for CT image formation. Comparisons to existing filter back-projection, iterative, and model-based reconstructions are now available in the literature. This review summarizes the prior reconstruction methods, introduces DLR, and then reviews recent findings from DLR from a physics and clinical perspective.

Recent Findings

DLR has been shown to allow for noise magnitude reductions relative to filtered back-projection without suffering from “plastic” or “blotchy” noise texture that was found objectionable with most iterative and model-based solutions. Clinically, early reader studies have reported increases in subjective quality scores and studies have successfully implemented DLR-enabled dose reductions.

Summary

The future of CT image reconstruction is bright; deep learning methods have only started to tackle problems in this space via addressing noise reduction. Artifact mitigation and spectral applications likely be future candidates for DLR applications.

Image quality in liver CT: low-dose deep learning vs standard-dose model-based iterative reconstructions

OBJECTIVES

To compare the overall image quality and detectability of significant (malignant and pre-malignant) liver lesions of low-dose liver CT (LDCT, 33.3% dose) using deep learning denoising (DLD) to standard-dose CT (SDCT, 100% dose) using model-based iterative reconstruction (MBIR).

METHODS

In this retrospective study, CT images of 80 patients with hepatic focal lesions were included. For noninferiority analysis of overall image quality, a margin of - 0.5 points (scored in a 5-point scale) for the difference between scan protocols was pre-defined. Other quantitative or qualitative image quality assessments were performed. Additionally, detectability of significant liver lesions was compared, with 64 pairs of CT, using the jackknife alternative free-response ROC analysis, with noninferior margin defined by the lower limit of 95% confidence interval (CI) of the difference of figure-of-merit less than - 0.1.

RESULTS

The mean overall image quality scores with LDCT and SDCT were 3.77 ± 0.38 and 3.94 ± 0.34, respectively, demonstrating a difference of - 0.17 (95% CI: - 0.21 to - 0.12), which did not cross the predefined noninferiority margin of - 0.5. Furthermore, LDCT showed significantly superior quantitative results of liver lesion contrast to noise ratio (p < 0.05). However, although LDCT scored higher than the average score in qualitative image quality assessments, they were significantly lower than those of SDCT (p < 0.05). Figure-of-merit for lesion detection was 0.859 for LDCT and 0.878 for SDCT, showing noninferiority (difference: - 0.019, 95% CI: - 0.058 to 0.021).

CONCLUSION

LDCT using DLD with 67% radiation dose reduction showed non-inferior overall image quality and lesion detectability, compared to SDCT.

KEY POINTS

• Low-dose liver CT using deep learning denoising (DLD), at 67% dose reduction, provided non-inferior overall image quality compared to standard-dose CT using model-based iterative reconstruction (MBIR). • Low-dose CT using DLD showed significantly less noise and higher CNR _{lesion to liver} than standard-dose CT using MBIR and demonstrated at least average image quality score among all readers, albeit with lower scores than standard-dose CT using MBIR. • Low-dose liver CT showed noninferior detectability for malignant and pre-malignant liver lesions, compared to standard-dose CT.

Reader Performance as a Function of Patient Size for the Detection of Hepatic Metastases

OBJECTIVE

To investigate reader performance as a function of patient size for the detection of hepatic metastases when an automatic exposure control (AEC) system is used, which varies image noise as a function of patient size.

METHODS

Abdominal computed tomograhy examinations across 100, 120, 160, and 200 quality reference tube current-time product were collected, involving a cohort of 83 patients. Three radiologists identified hepatic metastases across all dose levels. Partial Spearman rank correlation and multivariate logistic regression were used to evaluate correlations between reader performance and patient size and lesion size/contrast while accounting for potential confounding effects. Analyses were repeated on an emulated less-variable noise AEC.

RESULTS

No statistically significant correlation was observed between patient size and radiologist performance (for variable-noise AEC: range of partial Spearman ρ, -0.157 to -0.035]; range of adjusted odds ratios, 0.987, 1.006).

CONCLUSIONS

Reader performance was independent of patient size, suggesting that variable-noise AEC provides better modulation for larger patients than constant-noise AEC.

Added Value of Deep Learning-based Detection System for Multiple Major Findings on Chest Radiographs: A Randomized Crossover Study

Background Previous studies assessing the effects of computer-aided detection on observer performance in the reading of chest radiographs used a sequential reading design that may have biased the results because of reading order or recall bias. Purpose To compare observer performance in detecting and localizing major abnormal findings including nodules, consolidation, interstitial opacity, pleural effusion, and pneumothorax on chest radiographs without versus with deep learning-based detection (DLD) system assistance in a randomized crossover design. Materials and Methods This study included retrospectively collected normal and abnormal chest radiographs between January 2016 and December 2017 (https://cris.nih.go.kr/; registration no. KCT0004147). The radiographs were randomized into two groups, and six observers, including thoracic radiologists, interpreted each radiograph without and with use of a commercially available DLD system by using a crossover design with a washout period. Jackknife alternative free-response receiver operating characteristic (JAFROC) figure of merit (FOM), area under the receiver operating characteristic curve (AUC), sensitivity, specificity, false-positive findings per image, and reading times of observers with and without the DLD system were compared by using McNemar and paired t tests. Results A total of 114 normal (mean patient age ± standard deviation, 51 years ± 11; 58 men) and 114 abnormal (mean patient age, 60 years ± 15; 75 men) chest radiographs were evaluated. The radiographs were randomized to two groups: group A (n = 114) and group B (n = 114). Use of the DLD system improved the observers' JAFROC FOM (from 0.90 to 0.95, P = .002), AUC (from 0.93 to 0.98, P = .002), per-lesion sensitivity (from 83% [822 of 990 lesions] to 89.1% [882 of 990 lesions], P = .009), per-image sensitivity (from 80% [548 of 684 radiographs] to 89% [608 of 684 radiographs], P = .009), and specificity (from 89.3% [611 of 684 radiographs] to 96.6% [661 of 684 radiographs], P = .01) and reduced the reading time (from 10-65 seconds to 6-27 seconds, P < .001). The DLD system alone outperformed the pooled observers (JAFROC FOM: 0.96 vs 0.90, respectively, P = .007; AUC: 0.98 vs 0.93, P = .003). Conclusion Observers including thoracic radiologists showed improved performance in the detection and localization of major abnormal findings on chest radiographs and reduced reading time with use of a deep learning-based detection system. © RSNA, 2021 Online supplemental material is available for this article.

Low-dose CT image and projection dataset

PURPOSE

To describe a large, publicly available dataset comprising computed tomography (CT) projection data from patient exams, both at routine clinical doses and simulated lower doses.

ACQUISITION AND VALIDATION METHODS

The library was developed under local ethics committee approval. Projection and image data from 299 clinically performed patient CT exams were archived for three types of clinical exams: noncontrast head CT scans acquired for acute cognitive or motor deficit, low-dose noncontrast chest scans acquired to screen high-risk patients for pulmonary nodules, and contrast-enhanced CT scans of the abdomen acquired to look for metastatic liver lesions. Scans were performed on CT systems from two different CT manufacturers using routine clinical protocols. Projection data were validated by reconstructing the data using several different reconstruction algorithms and through use of the data in the 2016 Low Dose CT Grand Challenge. Reduced dose projection data were simulated for each scan using a validated noise-insertion method. Radiologists marked location and diagnosis for detected pathologies. Reference truth was obtained from the patient medical record, either from histology or subsequent imaging.

DATA FORMAT AND USAGE NOTES

Projection datasets were converted into the previously developed DICOM-CT-PD format, which is an extended DICOM format created to store CT projections and acquisition geometry in a nonproprietary format. Image data are stored in the standard DICOM image format and clinical data in a spreadsheet. Materials are provided to help investigators use the DICOM-CT-PD files, including a dictionary file, data reader, and user manual. The library is publicly available from The Cancer Imaging Archive (https://doi.org/10.7937/9npb-2637).

POTENTIAL APPLICATIONS

This CT data library will facilitate the development and validation of new CT reconstruction and/or denoising algorithms, including those associated with machine learning or artificial intelligence. The provided clinical information allows evaluation of task-based diagnostic performance.

Undetected Lung Cancer at Posteroanterior Chest Radiography: Potential Role of a Deep Learning-based Detection Algorithm

Purpose

To evaluate the performance of a deep learning-based algorithm in detecting lung cancers not reported on posteroanterior chest radiographs during routine practice.

Materials and Methods

The retrospective test dataset included 168 posteroanterior chest radiographs acquired between March 2017 and December 2018 (168 patients; mean age, 71.9 years ± 9.5 [standard deviation]; age range, 42-91 years) with 187 lung cancers (mean size, 2.3 cm ± 1.2) undetected during initial clinical evaluation, and 50 normal chest radiographs. CT served as the reference standard for ground truth. Four thoracic radiologists independently reevaluated the chest radiographs for lung nodules both without and with the aid of the algorithm. The performances of the algorithm and the radiologists were evaluated and compared on a per-chest radiograph basis and a per-lesion basis, according to the area under the receiver operating characteristic curve (AUROC) and area under the jackknife free-response ROC curve (AUFROC).

Results

The algorithm showed excellent diagnostic performances both in terms of per-chest radiograph classification (AUROC, 0.899) and per-lesion localization (AUFROC, 0.744); both of these values were significantly higher than those of the radiologists (AUROC, 0.634-0.663; AUFROC, 0.619-0.651; P < .001 for all). The algorithm also demonstrated higher sensitivity (69.6% [117 of 168] vs 47.0% [316 of 672]; P < .001) and specificity (94.0% [47 of 50] vs 78.0% [156 of 200]; P = .01). When assisted by the algorithm, the radiologists' AUROC (0.634-0.663 vs 0.685-0.724; P < 0.01 for all) and pooled AUFROC (0.636 vs 0.688; P = .03) substantially improved. The false-positive rate of the algorithm, that is, the total number of false-positive nodules divided by the total number of chest radiographs, was similar to that of pooled radiologists (21.1% [46 of 218] vs 19.0% [166 of 872]; P > .05).

Conclusion

A deep learning-based nodule detection algorithm showed excellent detection performance of lung cancers that were not reported on chest radiographs during routine practice and significantly reduced reading errors when used as a second reader.Supplemental material is available for this article.© RSNA, 2020See also commentary by White in this issue.

The image quality of deep-learning image reconstruction of chest CT images on a mediastinal window setting

AIM

To assess the image quality of deep-learning image reconstruction (DLIR) of chest computed tomography (CT) images on a mediastinal window setting in comparison to an adaptive statistical iterative reconstruction (ASiR-V).

MATERIALS AND METHODS

Thirty-six patients were evaluated retrospectively. All patients underwent contrast-enhanced chest CT and thin-section images were reconstructed using filtered back projection (FBP); ASiR-V (60% and 100% blending setting); and DLIR (low, medium, and high settings). Image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were evaluated objectively. Two independent radiologists evaluated ASiR-V 60% and DLIR subjectively, in comparison with FBP, on a five-point scale in terms of noise, streak artefact, lymph nodes, small vessels, and overall image quality on a mediastinal window setting (width 400 HU, level 60 HU). In addition, image texture of ASiR-Vs (60% and 100%) and DLIR-high was analysed subjectively.

RESULTS

Compared with ASiR-V 60%, DLIR-med and DLIR-high showed significantly less noise, higher SNR, and higher CNR (p<0.0001). DLIR-high and ASiR-V 100% were not significantly different regarding noise (p=0.2918) and CNR (p=0.0642). At a higher DLIR setting, noise was lower and SNR and CNR were higher (p<0.0001). DLIR-high showed the best subjective scores for noise, streak artefact, and overall image quality (p<0.0001). Compared with ASiR-V 60%, DLIR-med and DLIR-high scored worse in the assessment of small vessels (p<0.0001). The image texture of DLIR-high was significantly finer than that of ASIR-Vs (p<0.0001).

CONCLUSIONS

DLIR-high improved the objective parameters and subjective image quality by reducing noise and streak artefacts and providing finer image texture.

Extravalidation and reproducibility results of a commercial deep learning-based automatic detection algorithm for pulmonary nodules on chest radiographs at tertiary hospital

INTRODUCTION

To extra validate and evaluate the reproducibility of a commercial deep convolutional neural network (DCNN) algorithm for pulmonary nodules on chest radiographs (CRs) and to compare its performance with radiologists.

METHODS

This retrospective study enrolled 434 CRs (normal to abnormal ratio, 246:188) from 378 patients that visited a tertiary hospital. DCNN performance was compared with two radiology residents and two thoracic radiologists. Abnormality assessment (using the area under the receiver operating ch3cteristics (AUROC)) and nodule detection (using jackknife alternative free-response ROC (JAFROC)) were compared among three groups (DCNN only, radiologist without DCNN and radiologist with DCNN). A subset of 56 paired cases, having two CRs taken within a 7-day period, were assessed for intraobserver reproducibility using the intraclass correlation coefficient. Independent characteristics of pulmonary nodules detected by DCNN were assessed by multiple logistic regression analysis.

RESULTS

The AUROC for abnormality detection for the three groups were 0.87, 0.93 and 0.96, respectively (P < 0.05), whereas the JAFROC analysis of nodule detection was 0.926, 0.929 and 0.964. Reproducibility for the three groups was 0.80, 0.67 and 0.80, which shows an increase in radiologists using DCNN (P < 0.05). Nodules detected by DCNN were more solid, round-shaped and well marginated, not masked and laterally located (P < 0.05).

CONCLUSIONS

Extra validation results of DCNN showed high ROC results and there was a significant improvement in the performance when radiologists used DCNN. Reproducibility by DCNN alone showed good agreement, and there was an improvement from moderate to good agreement for radiologists using DCNN.

Observer Performance for Detection of Pulmonary Nodules at Chest CT over a Large Range of Radiation Dose Levels

Background There is a wide variation in radiation dose levels that can be used with chest CT in order to detect indeterminate pulmonary nodules. Purpose To compare the performance of lower-radiation-dose chest CT with that of routine dose in the detection of indeterminate pulmonary nodules 5 mm or greater. Materials and Methods In this retrospective study, CT projection data from 83 routine-dose chest CT examinations performed in 83 patients (120 kV, 70 quality reference mAs [QRM]) were collected between November 2013 and April 2014. Reference indeterminate pulmonary nodules were identified by two nonreader thoracic radiologists. By using validated noise insertion, five lower-dose data sets were reconstructed with filtered back projection (FBP) or iterative reconstruction (IR; 30 QRM with FBP, 10 QRM with IR, 5 QRM with FBP, 5 QRM with IR, and 2.5 QRM with IR). Three thoracic radiologists circled pulmonary nodules, rating confidence that the nodule was a 5-mm-or-greater indeterminate pulmonary nodule, and graded image quality. Analysis was performed on a per-nodule basis by using jackknife alternative free-response receiver operating characteristic figure of merit (FOM) and noninferiority limit of -0.10. Results There were 66 indeterminate pulmonary nodules (mean size, 8.6 mm ± 3.4 [standard deviation]; 21 part-solid nodules) in 42 patients (mean age, 51 years ± 17; 21 men and 21 women). Compared with the FOM for routine-dose CT (size-specific dose estimate, 6.5 mGy ± 1.8; FOM, 0.86 [95% confidence interval: 0.80, 0.91]), FOM was noninferior for all lower-dose configurations except for 2.5 QRM with IR. The sensitivity for subsolid nodules at 70 QRM was 60% (range, 48%-72%) and was significantly worse at a dose of 5 QRM and lower, whether or not IR was used (P < .05). Diagnostic image quality decreased with decreasing dose (P < .001) and was better with IR at 5 QRM (P < .05). Conclusion CT images reconstructed at dose levels down to 10 quality reference mAs (size-specific dose estimate, 0.9 mGy) had noninferior performance compared with routine dose in depicting pulmonary nodules. Iterative reconstruction improved subjective image quality but not performance at low dose levels. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by White and Kazerooni in this issue.

Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs

Background

Detection of active pulmonary tuberculosis on chest radiographs (CRs) is critical for the diagnosis and screening of tuberculosis. An automated system may help streamline the tuberculosis screening process and improve diagnostic performance.

Methods

We developed a deep learning–based automatic detection (DLAD) algorithm using 54c221 normal CRs and 6768 CRs with active pulmonary tuberculosis that were labeled and annotated by 13 board-certified radiologists. The performance of DLAD was validated using 6 external multicenter, multinational datasets. To compare the performances of DLAD with physicians, an observer performance test was conducted by 15 physicians including nonradiology physicians, board-certified radiologists, and thoracic radiologists. Image-wise classification and lesion-wise localization performances were measured using area under the receiver operating characteristic (ROC) curves and area under the alternative free-response ROC curves, respectively. Sensitivities and specificities of DLAD were calculated using 2 cutoffs (high sensitivity [98%] and high specificity [98%]) obtained through in-house validation.

Results

DLAD demonstrated classification performance of 0.977–1.000 and localization performance of 0.973–1.000. Sensitivities and specificities for classification were 94.3%–100% and 91.1%–100% using the high-sensitivity cutoff and 84.1%–99.0% and 99.1%–100% using the high-specificity cutoff. DLAD showed significantly higher performance in both classification (0.993 vs 0.746–0.971) and localization (0.993 vs 0.664–0.925) compared to all groups of physicians.

Conclusions

Our DLAD demonstrated excellent and consistent performance in the detection of active pulmonary tuberculosis on CR, outperforming physicians, including thoracic radiologists.

Can iterative reconstruction algorithms replace tube loading compensation in low kVp hepatic CT? Subjective versus objective image quality

Background

Hepatic computed tomography (CT) with decreased peak kilovoltage (kVp) may be used to reduce contrast medium doses in patients at risk of contrast-induced acute kidney injury (CI-AKI); however, it increases image noise. To preserve image quality, noise has been controlled by X-ray tube loading (mAs) compensation (TLC), i.e. increased mAs. Another option to control image noise would be to use iterative reconstructions (IR) algorithms without TLC (No-TLC). It is unclear whether this may preserve image quality or only reduce image noise.

Purpose

To evaluate image quality of 80 kVp hepatic CT with TLC and filtered back projection (FBP) compared with 80 kVp with No-TLC and IR algorithms (SAFIRE 3 and 5) in patients with eGFR <45 mL/min.

Material and Methods

Forty patients (BMI 18–32 kg/m²) were examined with both protocols following injection of 300 mg I/kg. Hepatic attenuation, image noise, enhancement, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and subjective image quality were evaluated for each patient.

Results

Comparing TLC/FBP with No-TLC/IR-S5, there were no significant differences regarding hepatic attenuation, image noise, enhancement, SNR and CNR: 114 vs. 115 HU, 14 vs. 14 HU, 55 vs. 57 HU, 8.0 vs. 8.4, and 3.8 vs. 4.0 in median, respectively. No-TLC/IR-S3 resulted in higher image noise and lower SNR and CNR than TLC/FBP. Subjective image quality scoring with visual grading showed statistically significantly inferior scores for IR-S5 images.

Conclusion

CT of 80 kVp to reduce contrast medium dose in patients at risk of CI-AKI combined with IR algorithms with unchanged tube loading to control image noise does not provide sufficient diagnostic quality.

High Acceleration Three-Dimensional T1-Weighted Dual Echo Dixon Hepatobiliary Phase Imaging Using Compressed Sensing-Sensitivity Encoding: Comparison of Image Quality and Solid Lesion Detectability with the Standard T1-Weighted Sequence

Objective

To compare a high acceleration three-dimensional (3D) T1-weighted gradient-recalled-echo (GRE) sequence using the combined compressed sensing (CS)-sensitivity encoding (SENSE) method with a conventional 3D GRE sequence using SENSE, with respect to image quality and detectability of solid focal liver lesions (FLLs) in the hepatobiliary phase (HBP) of gadoxetic acid-enhanced liver MRI.

Materials and Methods

A total of 217 patients with gadoxetic acid-enhanced liver MRI at 3T (54 in the preliminary study and 163 in the main study) were retrospectively included. In the main study, HBP imaging was done twice using the standard mDixon-3D-GRE technique with SENSE (acceleration factor [AF]: 2.8, standard mDixon-GRE) and the high acceleration mDixon-3D GRE technique using the combined CS-SENSE technique (CS-SENSE mDixon-GRE). Two abdominal radiologists assessed the two MRI data sets for image quality in consensus. Three other abdominal radiologists independently assessed the diagnostic performance of each data set and its ability to detect solid FLLs in 117 patients with 193 solid nodules and compared them using jackknife alternative free-response receiver operating characteristics (JAFROC).

Results

There was no significant difference in the overall image quality. CS-SENSE mDixon-GRE showed higher image noise, but lesser motion artifact levels compared with the standard mDixon-GRE (all p < 0.05). In terms of lesion detection, reader-averaged figures-of-merit estimated with JAFROC was 0.918 for standard mDixon-GRE, and 0.953 for CS-SENSE mDixon-GRE (p = 0.142). The non-inferiority of CS-SENSE mDixon-GRE over standard mDixon-GRE was confirmed (difference: 0.064 [−0.012, 0.081]).

Conclusion

The CS-SENSE mDixon-GRE HBP sequence provided comparable overall image quality and non-inferior solid FFL detectability compared with the standard mDixon-GRE sequence, with reduced acquisition time.

Evaluation of Lower-Dose Spiral Head CT for Detection of Intracranial Findings Causing Neurologic Deficits

BACKGROUND AND PURPOSE

Despite the frequent use of unenhanced head CT for the detection of acute neurologic deficit, the radiation dose for this exam varies widely. Our aim was to evaluate the performance of lower-dose head CT for detection of intracranial findings resulting in acute neurologic deficit.

MATERIALS AND METHODS

Projection data from 83 patients undergoing unenhanced spiral head CT for suspected neurologic deficits were collected. Cases positive for infarction, intra-axial hemorrhage, mass, or extra-axial hemorrhage required confirmation by histopathology, surgery, progression of findings, or corresponding neurologic deficit; cases negative for these target diagnoses required negative assessments by two neuroradiologists and a clinical neurologist. A routine dose head CT was obtained using 250 effective mAs and iterative reconstruction. Lower-dose configurations were reconstructed (25-effective mAs iterative reconstruction, 50-effective mAs filtered back-projection and iterative reconstruction, 100-effective mAs filtered back-projection and iterative reconstruction, 200-effective mAs filtered back-projection). Three neuroradiologists circled findings, indicating diagnosis, confidence (0-100), and image quality. The difference between the jackknife alternative free-response receiver operating characteristic figure of merit at routine and lower-dose configurations was estimated. A lower 95% CI estimate of the difference greater than -0.10 indicated noninferiority.

RESULTS

Forty-two of 83 patients had 70 intracranial findings (29 infarcts, 25 masses, 10 extra- and 6 intra-axial hemorrhages) at routine head CT (CT dose index = 38.3 mGy). The routine-dose jackknife alternative free-response receiver operating characteristic figure of merit was 0.87 (95% CI, 0.81-0.93). Noninferiority was shown for 100-effective mAs iterative reconstruction (figure of merit difference, -0.04; 95% CI, -0.08 to 0.004) and 200-effective mAs filtered back-projection (-0.02; 95% CI, -0.06 to 0.02) but not for 100-effective mAs filtered back-projection (-0.06; 95% CI, -0.10 to -0.02) or lower-dose levels. Image quality was better at higher-dose levels and with iterative reconstruction (P < .05).

CONCLUSIONS

Observer performance for dose levels using 100-200 eff mAs was noninferior to that observed at 250 effective mAs with iterative reconstruction, with iterative reconstruction preserving noninferiority at a mean CT dose index of 15.2 mGy.

Localization of liver lesions in abdominal CT imaging: I. Correlation of human observer performance between anatomical and uniform backgrounds

The purpose of this study was to determine the correlation between human observer performance for localization of small low contrast lesions within uniform water background versus an anatomical liver background, under the conditions of varying dose, lesion size, and reconstruction algorithm. Liver lesions (5 mm, 7 mm, and 9 mm, contrast:  -21 HU) were digitally inserted into CT projection data of ten normal patients in vessel-free liver regions. Noise was inserted into the projection data to create three image sets: full dose and simulated half and quarter doses. Images were reconstructed with a standard filtered back projection (FBP) and an iterative reconstruction (IR) algorithm. Lesion and noise insertion procedures were repeated with water phantom data. Two-dimensional regions of interest (66 lesion-present, 34 lesion-absent) were selected, randomized, and independently reviewed by three medical physicists to identify the most likely location of the lesion and provide a confidence score. Locations and confidence scores were assessed using the area under the localization receiver operating characteristic curve (Az_LROC). We examined the correlation between human performance for the liver and uniform water backgrounds as dose, lesion size, and reconstruction algorithm varied. As lesion size or dose increased, reader localization performance improved. For full dose IR images, the Az_LROC for 5, 7, and 9 mm lesions were 0.53, 0.91, and 0.97 (liver) and 0.51, 0.96, and 0.99 (uniform water), respectively. Similar trends were seen with other parameters. Performance values for liver and uniform backgrounds were highly correlated for both reconstruction algorithms, with a Spearman correlation of ρ  =  0.97, and an average difference in Az_LROC of 0.05  ±  0.04. For the task of localizing low contrast liver lesions, human observer performance was highly correlated between anatomical and uniform backgrounds, suggesting that lesion localization studies emulating a clinical test of liver lesion detection can be performed using a uniform background.

Can Realistic Liver Tissue Surrogates Accurately Quantify the Impact of Reduced-kV Imaging on Attenuation and Contrast of Parenchyma and Lesions?

RATIONALE AND OBJECTIVES

To assess if a liquid tissue surrogate for the liver (LTSL) can emulate contrast-enhanced liver parenchyma and lesions and quantify the impact of reduced-kV imaging as a function of lesion contrast, phase of enhancement, and phantom size.

MATERIALS AND METHODS

First, CT attenuation of LTSL- and water-iodine solutions were measured as a function of iodine concentration and tube potential. For each solution, the iodine concentration was determined to emulate liver parenchyma at 120 kV. CT attenuation for both solutions was predicted for different tube potentials and compared to published patient data. Second, liver parenchyma in late arterial phase (LA: +92 HU at 120 kV) and portal venous phase (PV: +112 HU at 120 kV) was emulated using LTSL-iodine and a two-size phantom. Fourteen setups of hyper- and hypoattenuating lesions (lesion-to-parenchyma contrast (C_LP) = -50 to +50HU) were created. Each combination of C_LP, phase, and size was imaged at 80, 100, 120, and 140 kV at constant radiation dose. CT attenuation, C_LP, and lesion-to-parenchyma contrast-to-noise ratio (CNR_LP) were assessed and compared to a theoretical model.

RESULTS

LTSL-iodine more accurately emulated the CT attenuation of liver parenchyma across different tube potentials compared to water-iodine solutions. The theoretical model was confirmed by the empirical measurements using LTSL-iodine solutions: attenuation, C_LP, and CNR_LP increased when the tube potential decreased (p < 0.001). This trend was independent of lesion contrast, phase, and size. The absolute improvement in C_LP and CNR_LP, however, was inversely related to the magnitude of C_LP at 140kV.

CONCLUSION

LTSL accurately emulated the energy-dependent CT attenuation characteristics of contrast-enhanced liver parenchyma and lesions. The relative improvement in C_LP and CNR_LP by applying reduced-kV imaging was independent of lesion contrast, phase, and size while the absolute improvement decreased for low-contrast lesions.

Development and Validation of a Deep Learning-Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs

IMPORTANCE

Interpretation of chest radiographs is a challenging task prone to errors, requiring expert readers. An automated system that can accurately classify chest radiographs may help streamline the clinical workflow.

OBJECTIVES

To develop a deep learning-based algorithm that can classify normal and abnormal results from chest radiographs with major thoracic diseases including pulmonary malignant neoplasm, active tuberculosis, pneumonia, and pneumothorax and to validate the algorithm's performance using independent data sets.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study developed a deep learning-based algorithm using single-center data collected between November 1, 2016, and January 31, 2017. The algorithm was externally validated with multicenter data collected between May 1 and July 31, 2018. A total of 54 221 chest radiographs with normal findings from 47 917 individuals (21 556 men and 26 361 women; mean [SD] age, 51 [16] years) and 35 613 chest radiographs with abnormal findings from 14 102 individuals (8373 men and 5729 women; mean [SD] age, 62 [15] years) were used to develop the algorithm. A total of 486 chest radiographs with normal results and 529 with abnormal results (1 from each participant; 628 men and 387 women; mean [SD] age, 53 [18] years) from 5 institutions were used for external validation. Fifteen physicians, including nonradiology physicians, board-certified radiologists, and thoracic radiologists, participated in observer performance testing. Data were analyzed in August 2018.

EXPOSURES

Deep learning-based algorithm.

MAIN OUTCOMES AND MEASURES

Image-wise classification performances measured by area under the receiver operating characteristic curve; lesion-wise localization performances measured by area under the alternative free-response receiver operating characteristic curve.

RESULTS

The algorithm demonstrated a median (range) area under the curve of 0.979 (0.973-1.000) for image-wise classification and 0.972 (0.923-0.985) for lesion-wise localization; the algorithm demonstrated significantly higher performance than all 3 physician groups in both image-wise classification (0.983 vs 0.814-0.932; all P < .005) and lesion-wise localization (0.985 vs 0.781-0.907; all P < .001). Significant improvements in both image-wise classification (0.814-0.932 to 0.904-0.958; all P < .005) and lesion-wise localization (0.781-0.907 to 0.873-0.938; all P < .001) were observed in all 3 physician groups with assistance of the algorithm.

CONCLUSIONS AND RELEVANCE

The algorithm consistently outperformed physicians, including thoracic radiologists, in the discrimination of chest radiographs with major thoracic diseases, demonstrating its potential to improve the quality and efficiency of clinical practice.

Computed Tomography Technology-and Dose-in the 21st Century

In the last decade or so, a number of disruptive technological advances have taken place in x-ray computed tomography, making possible new clinical applications. Changes in scanner design have included the use of two x-ray sources and two detectors or the use of large detector arrays that provide 16 cm of longitudinal coverage in one gantry rotation. These advances have allowed images of the entire heart to be acquired in just one heartbeat, lowering the effective dose from cardiac computed tomography from ~15 mSv to <1 mSv. Dual-energy computed tomography is now in widespread clinical use, enabling the assessment of material composition and concentration, as well as a range of new clinical applications. An emerging technology known as photon-counting detector computed tomography directly measures the energies of detected photons and is capable of simultaneously acquiring more than two energy data sets. Photon-counting detector computed tomography also provides advantages such as the ability to reject electronic noise, better iodine contrast-to-noise for a given dose, and spatial resolution as fine as 150 μm. Optimized x-ray tube potential selection has allowed reduction in radiation and contrast doses. Finally, wide adoption of iterative reconstruction and noise-reduction techniques has occurred. In all, body computed tomography doses have fallen dramatically, for example, by over a factor of 3 from the early 1980s. All of these advances increase the medical benefit and decrease the potential radiation risk associated with computed tomography. However, care must be taken to ensure that doses are not lowered to the level at which the clinical task is compromised.

Detection of Colorectal Hepatic Metastases Is Superior at Standard Radiation Dose CT versus Reduced Dose CT

Purpose To evaluate colorectal cancer hepatic metastasis detection and characterization between reduced radiation dose (RD) and standard dose (SD) contrast material-enhanced CT of the abdomen and to qualitatively compare between filtered back projection (FBP) and iterative reconstruction algorithms. Materials and Methods In this prospective study (from May 2017 through November 2017), 52 adults with biopsy-proven colorectal cancer and suspected hepatic metastases at baseline CT underwent two portal venous phase CT scans: SD and RD in the same breath hold. Three radiologists, blinded to examination details, performed detection and characterization of 2-15-mm lesions on the SD FBP and RD adaptive statistical iterative reconstruction (ASIR)-V 60% series images. Readers assessed overall image quality and lesions between SD FBP and seven different iterative reconstructions. Two nonblinded consensus reviewers established the reference standard using the picture archiving and communication system lesion marks of each reader, multiple comparison examinations, and clinical data. Results RD CT resulted in a mean dose reduction of 54% compared with SD. Of the 260 lesions (233 metastatic, 27 benign), 212 (82%; 95% confidence interval [CI]: 76%, 86%) were detected with RD CT, whereas 252 (97%; 95% CI: 94%, 99%) were detected with SD (P < .001); per-lesion sensitivity was 79% (95% CI: 74%, 84%) and 94% (95% CI: 90%, 96%) (P < .001), respectively. Mean qualitative scores ranked SD images as higher quality than RD series images, and ASIR-V ranked higher than ASIR and Veo 3.0. Conclusion CT evaluation of colorectal liver metastases is compromised with modest radiation dose reduction, and the use of iterative reconstructions could not maintain observer performance. © RSNA, 2018.

Development and Validation of Deep Learning-based Automatic Detection Algorithm for Malignant Pulmonary Nodules on Chest Radiographs

Purpose To develop and validate a deep learning-based automatic detection algorithm (DLAD) for malignant pulmonary nodules on chest radiographs and to compare its performance with physicians including thoracic radiologists. Materials and Methods For this retrospective study, DLAD was developed by using 43 292 chest radiographs (normal radiograph-to-nodule radiograph ratio, 34 067:9225) in 34 676 patients (healthy-to-nodule ratio, 30 784:3892; 19 230 men [mean age, 52.8 years; age range, 18-99 years]; 15 446 women [mean age, 52.3 years; age range, 18-98 years]) obtained between 2010 and 2015, which were labeled and partially annotated by 13 board-certified radiologists, in a convolutional neural network. Radiograph classification and nodule detection performances of DLAD were validated by using one internal and four external data sets from three South Korean hospitals and one U.S. hospital. For internal and external validation, radiograph classification and nodule detection performances of DLAD were evaluated by using the area under the receiver operating characteristic curve (AUROC) and jackknife alternative free-response receiver-operating characteristic (JAFROC) figure of merit (FOM), respectively. An observer performance test involving 18 physicians, including nine board-certified radiologists, was conducted by using one of the four external validation data sets. Performances of DLAD, physicians, and physicians assisted with DLAD were evaluated and compared. Results According to one internal and four external validation data sets, radiograph classification and nodule detection performances of DLAD were a range of 0.92-0.99 (AUROC) and 0.831-0.924 (JAFROC FOM), respectively. DLAD showed a higher AUROC and JAFROC FOM at the observer performance test than 17 of 18 and 15 of 18 physicians, respectively (P < .05), and all physicians showed improved nodule detection performances with DLAD (mean JAFROC FOM improvement, 0.043; range, 0.006-0.190; P < .05). Conclusion This deep learning-based automatic detection algorithm outperformed physicians in radiograph classification and nodule detection performance for malignant pulmonary nodules on chest radiographs, and it enhanced physicians' performances when used as a second reader. © RSNA, 2018 Online supplemental material is available for this article.

Observer Performance with Varying Radiation Dose and Reconstruction Methods for Detection of Hepatic Metastases

Purpose To estimate the ability of lower dose levels and iterative reconstruction (IR) to display hepatic metastases that can be detected by radiologists. Materials and Methods Projection data from 83 contrast agent-enhanced CT examinations were collected. Metastases were defined by histopathologic analysis or progression and regression. Lower radiation dose configurations were reconstructed at five dose levels with filtered back projection (FBP) and IR (automatic exposure control settings: 80, 100, 120, 160, and 200 quality reference mAs [QRM]). Three abdominal radiologists circumscribed metastases, indicating confidence (confidence range, 0-100) and image quality. Noninferiority was assessed by using jackknife alternative free-response receiver operating characteristic (JAFROC) analysis (noninferiority limit, -0.10) and reader agreement rules, which required identification of metastases identified at routine dose, and no nonlesion localizations in patients negative for metastases, in 71 or more patient CT examinations (of 83), for each configuration. Results There were 123 hepatic metastases (mean size, 1.4 cm; median volume CT dose index and size-specific dose estimate, 11.0 and 13.4 mGy, respectively). By using JAFROC figure of merit, 100 QRM FBP did not meet noninferiority criteria and had estimated performance difference from routine dose of -0.08 (95% confidence interval: -0.11, -0.04). Preset reader agreement rules were not met for 100 QRM IR or 80 QRM IR, but were met for doses 120 QRM or higher (ie, size-specific dose estimate ≥ 8.0 mGy). IR improved image quality (P < .05) but not reader performance. Other than 160 QRM IR, lower dose levels were associated with reduced confidence in metastasis detection (P < .001). Conclusion For detection of hepatic metastases by using contrast-enhanced CT, dose levels that corresponded to 120 quality reference mAs (size-specific dose estimate, 8.0 mGy) and higher performed similarly to 200 quality reference mAs with filtered back projection. © RSNA, 2018 Online supplemental material is available for this article.

Comparison of Full- and Half-Dose Image Reconstruction With Filtered Back Projection or Sinogram-Affirmed Iterative Reconstruction in Dual-Source Single-Energy MDCT Urography

OBJECTIVE

The purpose of this study is to prospectively compare the image quality of and confidence in the presence of a lesion on CT urography images acquired using filtered back projection (FBP) with 100% and 50% radiation doses with those for images simultaneously acquired using sinogram-affirmed iterative reconstruction with strength 3 (SAFIRE) with 50% and 25% radiation doses for patients with a high risk for urothelial carcinomas.

SUBJECTS AND METHODS

A total of 150 patients randomly underwent CT urography examinations performed using a dual-source single-energy scanner. After the radiation output of each tube was adjusted, datasets at three radiation dose levels were reconstructed using FBP and SAFIRE. Seven radiologists subjectively assessed image quality and confidence in the presence of a lesion for a total of 1200 datasets. Nonparametric methods for cluster data were used to estimate AUC values for variance methods on the basis of a noninferiority margin of 0.05.

RESULTS

The mean AUC value for image quality in SAFIRE with a 25% radiation dose was significantly lower than that of FBP with 100% radiation dose (p < 0.05 for all). The mean AUC values for the presence of a lesion were 0.907 and 0.894 for FBP, respectively, at 100% and 50% radiation doses, respectively, and 0.900 and 0.799 for SAFIRE at 50% and 25% radiation doses, respectively. However, the image quality of images acquired with SAFIRE at a 25% radiation dose was significantly inferior to that of images acquired with FBP at a 100% radiation dose.

CONCLUSION

Regardless of the experience of the radiologist, CT urography images acquired with FBP and SAFIRE with a 50% radiation dose were noninferior to those acquired with FBP with a 100% radiation dose in terms of image quality and confidence in the presence of a lesion, whereas those acquired with SAFIRE with 25% radiation dose were inferior.

CT Detectability of Small Low-Contrast Hypoattenuating Focal Lesions: Iterative Reconstructions versus Filtered Back Projection

Purpose To investigate performance in detectability of small (≤1 cm) low-contrast hypoattenuating focal lesions by using filtered back projection (FBP) and iterative reconstruction (IR) algorithms from two major CT vendors across a range of 11 radiation exposures. Materials and Methods A low-contrast detectability phantom consisting of 21 low-contrast hypoattenuating focal objects (seven sizes between 2.4 and 10.0 mm, three contrast levels) embedded into a liver-equivalent background was scanned at 11 radiation exposures (volume CT dose index range, 0.5-18.0 mGy; size-specific dose estimate [SSDE] range, 0.8-30.6 mGy) with four high-end CT platforms. Data sets were reconstructed by using FBP and varied strengths of image-based, model-based, and hybrid IRs. Sixteen observers evaluated all data sets for lesion detectability by using a two-alternative-forced-choice (2AFC) paradigm. Diagnostic performances were evaluated by calculating area under the receiver operating characteristic curve (AUC) and by performing noninferiority analyses. Results At benchmark exposure, FBP yielded a mean AUC of 0.79 ± 0.09 (standard deviation) across all platforms which, on average, was approximately 2% lower than that observed with the different IR algorithms, which showed an average AUC of 0.81 ± 0.09 (P = .12). Radiation decreases of 30%, 50%, and 80% resulted in similar declines of observer detectability with FBP (mean AUC decrease, -0.02 ± 0.05, -0.03 ± 0.05, and -0.05 ± 0.05, respectively) and all IR methods investigated (mean AUC decrease, -0.00 ± 0.05, -0.04 ± 0.05, and -0.04 ± 0.05, respectively). For each radiation level and CT platform, variance in performance across observers was greater than that across reconstruction algorithms (P = .03). Conclusion Iterative reconstruction algorithms have limited radiation optimization potential in detectability of small low-contrast hypoattenuating focal lesions. This task may be further complicated by a high degree of variation in radiologists' performances, seemingly exceeding real performance differences among reconstruction algorithms. © RSNA, 2018 Online supplemental material is available for this article.

Low kV versus dual-energy virtual monoenergetic CT imaging for proven liver lesions: what are the advantages and trade-offs in conspicuity and image quality? A pilot study

PURPOSE

Single-energy low tube potential (SE-LTP) and dual-energy virtual monoenergetic (DE-VM) CT images both increase the conspicuity of hepatic lesions by increasing iodine signal. Our purpose was to compare the conspicuity of proven liver lesions, artifacts, and radiologist preferences in dose-matched SE-LTP and DE-VM images.

METHODS

Thirty-one patients with 72 proven liver lesions (21 benign, 51 malignant) underwent full-dose contrast-enhanced dual-energy CT (DECT). Half-dose images were obtained using single tube reconstruction of the dual-source SE-LTP projection data (80 or 100 kV), and by inserting noise into dual-energy projection data, with DE-VM images reconstructed from 40 to 70 keV. Three blinded gastrointestinal radiologists evaluated half-dose SE-LTP and DE-VM images, ranking and grading liver lesion conspicuity and diagnostic confidence (4-point scale) on a per-lesion basis. Image quality (noise, artifacts, sharpness) was evaluated, and overall image preference was ranked on per-patient basis. Lesion-to-liver contrast-to-noise ratio (CNR) was compared between techniques.

RESULTS

Mean lesion size was 1.5 ± 1.2 cm. Across the readers, the mean conspicuity ratings for 40, 45, and 50 keV half-dose DE-VM images were superior compared to other half-dose image sets (p < 0.0001). Per-lesion diagnostic confidence was similar between half-dose SE-LTP compared to half-dose DE-VM images (p ≥ 0.05; 1.19 vs. 1.24-1.32). However, SE-LTP images had less noise and artifacts and were sharper compared to DE-VM images less than 70 keV (p < 0.05). On a per-patient basis, radiologists preferred SE-LTP images the most and preferred 40-50 keV the least (p < 0.0001). Lesion CNR was also higher in SE-LTP images than DE-VM images (p < 0.01).

CONCLUSION

For the same applied dose level, liver lesions were more conspicuous using DE-VM compared to SE-LTP; however, SE-LTP images were preferred more than any single DE-VM energy level, likely due to lower noise and artifacts.

Low-dose CT for the detection and classification of metastatic liver lesions: Results of the 2016 Low Dose CT Grand Challenge

PURPOSE

Using common datasets, to estimate and compare the diagnostic performance of image-based denoising techniques or iterative reconstruction algorithms for the task of detecting hepatic metastases.

METHODS

Datasets from contrast-enhanced CT scans of the liver were provided to participants in an NIH-, AAPM- and Mayo Clinic-sponsored Low Dose CT Grand Challenge. Training data included full-dose and quarter-dose scans of the ACR CT accreditation phantom and 10 patient examinations; both images and projections were provided in the training data. Projection data were supplied in a vendor-neutral standardized format (DICOM-CT-PD). Twenty quarter-dose patient datasets were provided to each participant for testing the performance of their technique. Images were provided to sites intending to perform denoising in the image domain. Fully preprocessed projection data and statistical noise maps were provided to sites intending to perform iterative reconstruction. Upon return of the denoised or iteratively reconstructed quarter-dose images, randomized, blinded evaluation of the cases was performed using a Latin Square study design by 11 senior radiology residents or fellows, who marked the locations of identified hepatic metastases. Markings were scored against reference locations of clinically or pathologically demonstrated metastases to determine a per-lesion normalized score and a per-case normalized score (a faculty abdominal radiologist established the reference location using clinical and pathological information). Scores increased for correct detections; scores decreased for missed or incorrect detections. The winner for the competition was the entry that produced the highest total score (mean of the per-lesion and per-case normalized score). Reader confidence was used to compute a Jackknife alternative free-response receiver operating characteristic (JAFROC) figure of merit, which was used for breaking ties.

RESULTS

103 participants from 90 sites and 26 countries registered to participate. Training data were shared with 77 sites that completed the data sharing agreements. Subsequently, 41 sites downloaded the 20 test cases, which included only the 25% dose data (CTDIvol = 3.0 ± 1.8 mGy, SSDE = 3.5 ± 1.3 mGy). 22 sites submitted results for evaluation. One site provided binary images and one site provided images with severe artifacts; cases from these sites were excluded from review and the participants removed from the challenge. The mean (range) per-lesion and per-case normalized scores were -24.2% (-75.8%, 3%) and 47% (10%, 70%), respectively. Compared to reader results for commercially reconstructed quarter-dose images with no noise reduction, 11 of the 20 sites showed a numeric improvement in the mean JAFROC figure of merit. Notably two sites performed comparably to the reader results for full-dose commercial images. The study was not designed for these comparisons, so wide confidence intervals surrounded these figures of merit and the results should be used only to motivate future testing.

CONCLUSION

Infrastructure and methodology were developed to rapidly estimate observer performance for liver metastasis detection in low-dose CT examinations of the liver after either image-based denoising or iterative reconstruction. The results demonstrated large differences in detection and classification performance between noise reduction methods, although the majority of methods provided some improvement in performance relative to the commercial quarter-dose images with no noise reduction applied.

Influence of CT acquisition and reconstruction parameters on radiomic feature reproducibility

High-dimensional imaging features extracted from diagnostic imaging, called radiomics, are increasingly reported for diagnosis, prognosis, and response to therapy. Establishing the sensitivity of radiomic features to variation in scan protocols is necessary because acquisition and reconstruction parameters can vary widely across and within institutions. Our objective was to assess the reproducibility of radiomic features derived from computed tomography (CT) images by varying tube current (mA), noise index, and reconstruction [adaptive statistical iterative reconstruction (ASiR)], parameters increasingly varied by institutions seeking to reduce radiation dose in their patients. We extracted radiomic features from CT images of a uniform water phantom, anthropomorphic phantom, and a human scan. Scans were acquired from the phantoms with six tube currents (50, 100, 200, 300, 400, and 500 mA) and five noise index levels (12, 14, 16, 18, and 20), respectively. Scans of the phantoms and patient were reconstructed from 0% ASiR (i.e., filtered back projection) to 100% ASiR in increments of 10%. Two hundred and forty-eight well-known radiomic features were extracted from all scans. The concordance correlation coefficient was used to assess agreement of features. Our analysis suggests that image acquisition parameters (tube current, noise index) as well as the reconstruction technique strongly influence radiomic feature reproducibility and demonstrate a subset of reproducible features potentially usable in clinical practice.

Use of a channelized Hotelling observer to assess CT image quality and optimize dose reduction for iteratively reconstructed images

The use of iterative reconstruction (IR) algorithms in CT generally decreases image noise and enables dose reduction. However, the amount of dose reduction possible using IR without sacrificing diagnostic performance is difficult to assess with conventional image quality metrics. Through this investigation, achievable dose reduction using a commercially available IR algorithm without loss of low contrast spatial resolution was determined with a channelized Hotelling observer (CHO) model and used to optimize a clinical abdomen/pelvis exam protocol. A phantom containing 21 low contrast disks-three different contrast levels and seven different diameters-was imaged at different dose levels. Images were created with filtered backprojection (FBP) and IR. The CHO was tasked with detecting the low contrast disks. CHO performance indicated dose could be reduced by 22% to 25% without compromising low contrast detectability (as compared to full-dose FBP images) whereas 50% or more dose reduction significantly reduced detection performance. Importantly, default settings for the scanner and protocol investigated reduced dose by upward of 75%. Subsequently, CHO-based protocol changes to the default protocol yielded images of higher quality and doses more consistent with values from a larger, dose-optimized scanner fleet. CHO assessment provided objective data to successfully optimize a clinical CT acquisition protocol.

Estimation of Observer Performance for Reduced Radiation Dose Levels in CT: Eliminating Reduced Dose Levels That Are Too Low Is the First Step

RATIONALE AND OBJECTIVES

This study aims to estimate observer performance for a range of dose levels for common computed tomography (CT) examinations (detection of liver metastases or pulmonary nodules, and cause of neurologic deficit) to prioritize noninferior dose levels for further analysis.

MATERIALS AND METHODS

Using CT data from 131 examinations (abdominal CT, 44; chest CT, 44; head CT, 43), CT images corresponding to 4%-100% of the routine clinical dose were reconstructed with filtered back projection or iterative reconstruction. Radiologists evaluated CT images, marking specified targets, providing confidence scores, and grading image quality. Noninferiority was assessed using reference standards, reader agreement rules, and jackknife alternative free-response receiver operating characteristic figures of merit. Reader agreement required that a majority of readers at lower dose identify target lesions seen by the majority of readers at routine dose.

RESULTS

Reader agreement identified dose levels lower than 50% and 4% to have inadequate performance for detection of hepatic metastases and pulmonary nodules, respectively, but could not exclude any low dose levels for head CT. Estimated differences in jackknife alternative free-response receiver operating characteristic figures of merit between routine and lower dose configurations found that only the lowest dose configurations tested (ie, 30%, 4%, and 10% of routine dose levels for abdominal, chest, and head CT examinations, respectively) did not meet criteria for noninferiority. At lower doses, subjective image quality declined before observer performance. Iterative reconstruction was only beneficial when filtered back projection did not result in noninferior performance.

CONCLUSION

Opportunity exists for substantial radiation dose reduction using existing CT technology for common diagnostic tasks.

Evaluation of a projection-domain lung nodule insertion technique in thoracic computed tomography

Task-based assessment of computed tomography (CT) image quality requires a large number of cases with ground truth. Prospective case acquisition can be time-consuming. Inserting lesions into existing cases to simulate positive cases is a promising alternative. The aim was to evaluate a recently developed projection-based lesion insertion technique in thoracic CT. In total, 32 lung nodules of various attenuations were segmented from 21 patient cases, forward projected, inserted into projections, and reconstructed. Two experienced radiologists and two residents independently evaluated these nodules in two substudies. First, the 32 inserted and the 32 original nodules were presented in a randomized order and each received a score from 1 to 10 (1 = absolutely artificial to 10 = absolutely realistic). Second, the inserted and the corresponding original lesions were presented side-by-side to each reader. For the randomized evaluation, discrimination of real versus inserted nodules was poor with areas under the receiver operative characteristic curves being 0.57 [95% confidence interval (CI): 0.46 to 0.68], 0.69 (95% CI: 0.58 to 0.78), and 0.62 (95% CI: 0.54 to 0.69) for the two residents, two radiologists, and all four readers, respectively. Our projection-based lung nodule insertion technique provides a robust method to artificially generate positive cases that prove to be difficult to differentiate from real cases.

Effect of Radiation Dose Reduction and Reconstruction Algorithm on Image Noise, Contrast, Resolution, and Detectability of Subtle Hypoattenuating Liver Lesions at Multidetector CT: Filtered Back Projection versus a Commercial Model-based Iterative Reconstruction Algorithm

Purpose To determine the effect of radiation dose and iterative reconstruction (IR) on noise, contrast, resolution, and observer-based detectability of subtle hypoattenuating liver lesions and to estimate the dose reduction potential of the IR algorithm in question. Materials and Methods This prospective, single-center, HIPAA-compliant study was approved by the institutional review board. A dual-source computed tomography (CT) system was used to reconstruct CT projection data from 21 patients into six radiation dose levels (12.5%, 25%, 37.5%, 50%, 75%, and 100%) on the basis of two CT acquisitions. A series of virtual liver lesions (five per patient, 105 total, lesion-to-liver prereconstruction contrast of -15 HU, 12-mm diameter) were inserted into the raw CT projection data and images were reconstructed with filtered back projection (FBP) (B31f kernel) and sinogram-affirmed IR (SAFIRE) (I31f-5 kernel). Image noise (pixel standard deviation), lesion contrast (after reconstruction), lesion boundary sharpness (average normalized gradient at lesion boundary), and contrast-to-noise ratio (CNR) were compared. Next, a two-alternative forced choice perception experiment was performed (16 readers [six radiologists, 10 medical physicists]). A linear mixed-effects statistical model was used to compare detection accuracy between FBP and SAFIRE and to estimate the radiation dose reduction potential of SAFIRE. Results Compared with FBP, SAFIRE reduced noise by a mean of 53% ± 5, lesion contrast by 12% ± 4, and lesion sharpness by 13% ± 10 but increased CNR by 89% ± 19. Detection accuracy was 2% higher on average with SAFIRE than with FBP (P = .03), which translated into an estimated radiation dose reduction potential (±95% confidence interval) of 16% ± 13. Conclusion SAFIRE increases detectability at a given radiation dose (approximately 2% increase in detection accuracy) and allows for imaging at reduced radiation dose (16% ± 13), while maintaining low-contrast detectability of subtle hypoattenuating focal liver lesions. This estimated dose reduction is somewhat smaller than that suggested by past studies. ^© RSNA, 2017 Online supplemental material is available for this article.

Dual-Source Single-Energy Multidetector CT Used to Obtain Multiple Radiation Exposure Levels within the Same Patient: Phantom Development and Clinical Validation

Purpose To develop, in a phantom environment, a method to obtain multidetector computed tomographic (CT) data sets at multiple radiation exposure levels within the same patient and to validate its use for potential dose reduction by using different image reconstruction algorithms for the detection of liver metastases. Materials and Methods The American College of Radiology CT accreditation phantom was scanned by using a dual-source multidetector CT platform. By adjusting the radiation output of each tube, data sets at six radiation exposure levels (100%, 75%, 50%, 37.5%, 25%, and 12.5%) were reconstructed from two consecutive dual-source single-energy (DSSE) acquisitions, as well as a conventional single-source acquisition. A prospective, HIPAA-compliant, institutional review board-approved study was performed by using the same DSSE strategy in 19 patients who underwent multidetector CT of the liver for metastatic colorectal cancer. All images were reconstructed by using conventional weighted filtered back projection (FBP) and sinogram-affirmed iterative reconstruction with strength level of 3 (SAFIRE-3). Objective image quality metrics were compared in the phantom experiment by using multiple linear regression analysis. Generalized linear mixed-effects models were used to analyze image quality metrics and diagnostic performance for lesion detection by readers. Results The phantom experiment showed comparable image quality between DSSE and conventional single-source acquisition. In the patient study, the mean size-specific dose estimates for the six radiation exposure levels were 13.0, 9.8, 5.8, 4.4, 3.2, and 1.4 mGy. For each radiation exposure level, readers' perception of image quality and lesion conspicuity was consistently ranked superior with SAFIRE-3 when compared with FBP (P ≤ .05 for all comparisons). Reduction of up to 62.5% in radiation exposure by using SAFIRE-3 yielded similar reader rankings of image quality and lesion conspicuity when compared with routine-dose FBP. Conclusion A method was developed and validated to synthesize multidetector CT data sets at multiple radiation exposure levels within the same patient. This technique may provide a foundation for future clinical trials aimed at estimating potential radiation dose reduction by using iterative reconstructions. ^© RSNA, 2016 Online supplemental material is available for this article.

Can conclusions drawn from phantom-based image noise assessments be generalized to in vivo studies for the nonlinear model-based iterative reconstruction method?

PURPOSE

Phantom-based objective image quality assessment methods are widely used in the medical physics community. For a filtered backprojection (FBP) reconstruction-based linear or quasilinear imaging system, the use of this methodology is well justified. Many key image quality metrics acquired with phantom studies can be directly applied to in vivo human subject studies. Recently, a variety of image quality metrics have been investigated for model-based iterative image reconstruction (MBIR) methods and several novel characteristics have been discovered in phantom studies. However, the following question remains unanswered: can certain results obtained from phantom studies be generalized to in vivo animal studies and human subject studies? The purpose of this paper is to address this question.

METHODS

One of the most striking results obtained from phantom studies is a novel power-law relationship between noise variance of MBIR (σ(2)) and tube current-rotation time product (mAs): σ(2) ∝ (mAs)(-0.4) [K. Li et al., "Statistical model based iterative reconstruction (MBIR) in clinical CT systems: Experimental assessment of noise performance," Med. Phys. 41, 041906 (15pp.) (2014)]. To examine whether the same power-law works for in vivo cases, experimental data from two types of in vivo studies were analyzed in this paper. All scans were performed with a 64-slice diagnostic CT scanner (Discovery CT750 HD, GE Healthcare) and reconstructed with both FBP and a MBIR method (Veo, GE Healthcare). An Institutional Animal Care and Use Committee-approved in vivo animal study was performed with an adult swine at six mAs levels (10-290). Additionally, human subject data (a total of 110 subjects) acquired from an IRB-approved clinical trial were analyzed. In this clinical trial, a reduced-mAs scan was performed immediately following the standard mAs scan; the specific mAs used for the two scans varied across human subjects and were determined based on patient size and clinical indications. The measurements of σ(2) were performed at different mAs by drawing regions-of-interest (ROIs) in the liver and the subcutaneous fat. By applying a linear least-squares regression, the β values in the power-law relationship σ(2) ∝ (mAs)(-β) were measured for the in vivo data and compared with the value found in phantom experiments.

RESULTS

For the in vivo swine study, an exponent of β = 0.43 was found for MBIR, and the coefficient of determination (R(2)) for the corresponding least-squares power-law regression was 0.971. As a reference, the β and R(2) values for FBP were found to be 0.98 and 0.997, respectively, from the same study, which are consistent with the well-known σ(2) ∝ (mAs)(-1.0) relationship for linear CT systems. For the human subject study, the measured β values for the MBIR images were 0.41 ± 0.12 in the liver and 0.37 ± 0.12 in subcutaneous fat. In comparison, the β values for the FBP images were 1.04 ± 0.10 in the liver and 0.97 ± 0.12 in subcutaneous fat. The β values of MBIR and FBP obtained from the in vivo studies were found to be statistically equivalent to the corresponding β values from the phantom study within an equivalency interval of [ - 0.1, 0.1] (p < 0.05); across MBIR and FBP, the difference in β was statistically significant (p < 0.05).

CONCLUSIONS

Despite the nonlinear nature of the MBIR method, the power-law relationship, σ(2) ∝ (mAs)(-0.4), found from phantom studies can be applied to in vivo animal and human subject studies.

Evaluation of a projection-domain lung nodule insertion technique in thoracic CT

Task-based assessment of computed tomography (CT) image quality requires a large number of cases with ground truth. Inserting lesions into existing cases to simulate positive cases is a promising alternative approach. The aim of this study was to evaluate a recently-developed raw-data based lesion insertion technique in thoracic CT. Lung lesions were segmented from patient CT images, forward projected, and reinserted into the same patient CT projection data. In total, 32 nodules of various attenuations were segmented from 21 CT cases. Two experienced radiologists and 2 residents blinded to the process independently evaluated these inserted nodules in two sub-studies. First, the 32 inserted and the 32 original nodules were presented in a randomized order and each received a rating score from 1 to 10 (1=absolutely artificial to 10=absolutely realistic). Second, the inserted and the corresponding original lesions were presented side-by-side to each reader, who identified the inserted lesion and provided a confidence score (1=no confidence to 5=completely certain). For the randomized evaluation, discrimination of real versus artificial nodules was poor with areas under the receiver operative characteristic curves being 0.69 (95% CI: 0.58-0.78), 0.57 (95% CI: 0.46-0.68), and 0.62 (95% CI: 0.54-0.69) for the 2 radiologists, 2 residents, and all 4 readers, respectively. For the side-by-side evaluation, although all 4 readers correctly identified inserted lesions in 103/128 pairs, the confidence score was moderate (2.6). Our projection-domain based lung nodule insertion technique provides a robust method to artificially generate clinical cases that prove to be difficult to differentiate from real cases.

Prospective Evaluation of Reduced Dose Computed Tomography for the Detection of Low-Contrast Liver Lesions: Direct Comparison with Concurrent Standard Dose Imaging

OBJECTIVES

To prospectively compare the diagnostic performance of reduced-dose (RD) contrast-enhanced CT (CECT) with standard-dose (SD) CECT for detection of low-contrast liver lesions.

METHODS

Seventy adults with non-liver primary malignancies underwent abdominal SD-CECT immediately followed by RD-CECT, aggressively targeted at 60-70 % dose reduction. SD series were reconstructed using FBP. RD series were reconstructed with FBP, ASIR, and MBIR (Veo). Three readers-blinded to clinical history and comparison studies-reviewed all series, identifying liver lesions ≥4 mm. Non-blinded review by two experienced abdominal radiologists-assessing SD against available clinical and radiologic information-established the reference standard.

RESULTS

RD-CECT mean effective dose was 2.01 ± 1.36 mSv (median, 1.71), a 64.1 ± 8.8 % reduction. Pooled per-patient performance data were (sensitivity/specificity/PPV/NPV/accuracy) 0.91/0.78/0.60/0.96/0.81 for SD-FBP compared with RD-FBP 0.79/0.75/0.54/0.91/0.76; RD-ASIR 0.84/0.75/0.56/0.93/0.78; and RD-MBIR 0.84/0.68/0.49/0.92/0.72. ROC AUC values were 0.896/0.834/0.858/0.854 for SD-FBP/RD-FBP/RD-ASIR/RD-MBIR, respectively. RD-FBP (P = 0.002) and RD-MBIR (P = 0.032) AUCs were significantly lower than those of SD-FBP; RD-ASIR was not (P = 0.052). Reader confidence was lower for all RD series (P < 0.001) compared with SD-FBP, especially when calling patients entirely negative.

CONCLUSIONS

Aggressive CT dose reduction resulted in inferior diagnostic performance and reader confidence for detection of low-contrast liver lesions compared to SD. Relative to RD-ASIR, RD-FBP showed decreased sensitivity and RD-MBIR showed decreased specificity.

KEY POINTS

• Reduced-dose CECT demonstrates inferior diagnostic performance for detecting low-contrast liver lesions. • Reader confidence is lower with reduced-dose CECT compared to standard-dose CECT. • Overly aggressive dose reduction may result in misdiagnosis, regardless of reconstruction algorithm. • Careful consideration of perceived risks versus benefits of dose reduction is crucial.

Impact of number of repeated scans on model observer performance for a low-contrast detection task in computed tomography

Channelized Hotelling observer (CHO) models have been shown to correlate well with human observers for several phantom-based detection/classification tasks in clinical computed tomography (CT). A large number of repeated scans were used to achieve an accurate estimate of the model's template. The purpose of this study is to investigate how the experimental and CHO model parameters affect the minimum required number of repeated scans. A phantom containing 21 low-contrast objects was scanned on a 128-slice CT scanner at three dose levels. Each scan was repeated 100 times. For each experimental configuration, the low-contrast detectability, quantified as the area under receiver operating characteristic curve, [Formula: see text], was calculated using a previously validated CHO with randomly selected subsets of scans, ranging from 10 to 100. Using [Formula: see text] from the 100 scans as the reference, the accuracy from a smaller number of scans was determined. Our results demonstrated that the minimum number of repeated scans increased when the radiation dose level decreased, object size and contrast level decreased, and the number of channels increased. As a general trend, it increased as the low-contrast detectability decreased. This study provides a basis for the experimental design of task-based image quality assessment in clinical CT using CHO.

Radiation Dose Reduction in Pediatric Body CT Using Iterative Reconstruction and a Novel Image-Based Denoising Method

OBJECTIVE

The objective of this study was to evaluate the radiation dose reduction potential of a novel image-based denoising technique in pediatric abdominopelvic and chest CT examinations and compare it with a commercial iterative reconstruction method.

MATERIALS AND METHODS

Data were retrospectively collected from 50 (25 abdominopelvic and 25 chest) clinically indicated pediatric CT examinations. For each examination, a validated noise-insertion tool was used to simulate half-dose data, which were reconstructed using filtered back-projection (FBP) and sinogram-affirmed iterative reconstruction (SAFIRE) methods. A newly developed denoising technique, adaptive nonlocal means (aNLM), was also applied. For each of the 50 patients, three pediatric radiologists evaluated four datasets: full dose plus FBP, half dose plus FBP, half dose plus SAFIRE, and half dose plus aNLM. For each examination, the order of preference for the four datasets was ranked. The organ-specific diagnosis and diagnostic confidence for five primary organs were recorded.

RESULTS

The mean (± SD) volume CT dose index for the full-dose scan was 5.3 ± 2.1 mGy for abdominopelvic examinations and 2.4 ± 1.1 mGy for chest examinations. For abdominopelvic examinations, there was no statistically significant difference between the half dose plus aNLM dataset and the full dose plus FBP dataset (3.6 ± 1.0 vs 3.6 ± 0.9, respectively; p = 0.52), and aNLM performed better than SAFIRE. For chest examinations, there was no statistically significant difference between the half dose plus SAFIRE and the full dose plus FBP (4.1 ± 0.6 vs 4.2 ± 0.6, respectively; p = 0.67), and SAFIRE performed better than aNLM. For all organs, there was more than 85% agreement in organ-specific diagnosis among the three half-dose configurations and the full dose plus FBP configuration.

CONCLUSION

Although a novel image-based denoising technique performed better than a commercial iterative reconstruction method in pediatric abdominopelvic CT examinations, it performed worse in pediatric chest CT examinations. A 50% dose reduction can be achieved while maintaining diagnostic quality.

Technical Note: Measuring contrast- and noise-dependent spatial resolution of an iterative reconstruction method in CT using ensemble averaging

PURPOSE

The spatial resolution of iterative reconstruction (IR) in computed tomography (CT) is contrast- and noise-dependent because of the nonlinear regularization. Due to the severe noise contamination, it is challenging to perform precise spatial-resolution measurements at very low-contrast levels. The purpose of this study was to measure the spatial resolution of a commercially available IR method using ensemble-averaged images acquired from repeated scans.

METHODS

A low-contrast phantom containing three rods (7, 14, and 21 HU below background) was scanned on a 128-slice CT scanner at three dose levels (CTDIvol = 16, 8, and 4 mGy). Images were reconstructed using two filtered-backprojection (FBP) kernels (B40 and B20) and a commercial IR method (sinogram affirmed iterative reconstruction, SAFIRE, Siemens Healthcare) with two strength settings (I40-3 and I40-5). The same scan was repeated 100 times at each dose level. The modulation transfer function (MTF) was calculated based on the edge profile measured on the ensemble-averaged images.

RESULTS

The spatial resolution of the two FBP kernels, B40 and B20, remained relatively constant across contrast and dose levels. However, the spatial resolution of the two IR kernels degraded relative to FBP as contrast or dose level decreased. For a given dose level at 16 mGy, the MTF50% value normalized to the B40 kernel decreased from 98.4% at 21 HU to 88.5% at 7 HU for I40-3 and from 97.6% to 82.1% for I40-5. At 21 HU, the relative MTF50% value decreased from 98.4% at 16 mGy to 90.7% at 4 mGy for I40-3 and from 97.6% to 85.6% for I40-5.

CONCLUSIONS

A simple technique using ensemble averaging from repeated CT scans can be used to measure the spatial resolution of IR techniques in CT at very low contrast levels. The evaluated IR method degraded the spatial resolution at low contrast and high noise levels.