Radiomics in cone-beam breast CT for the prediction of axillary lymph node metastasis in breast cancer: a multi-center multi-device study.

OBJECTIVES: To develop a radiomics model in contrast-enhanced cone-beam breast CT (CE-CBBCT) for preoperative prediction of axillary lymph node (ALN) status and metastatic burden of breast cancer. METHODS: Two hundred and seventy-four patients who underwent CE-CBBCT examination with two scanners between 2012 and 2021 from two institutions were enrolled. The primary tumor was annotated in each patient image, from which 1781 radiomics features were extracted with PyRadiomics. After feature selection, support vector machine models were developed to predict ALN status and metastatic burden. To avoid overfitting on a specific patient subset, 100 randomly stratified splits were made to assign the patients to either training/fine-tuning or test set. Area under the receiver operating characteristic curve (AUC) of these radiomics models was compared to those obtained when training the models only with clinical features and combined clinical-radiomics descriptors. Ground truth was established by histopathology. RESULTS: One hundred and eighteen patients had ALN metastasis (N + (≥ 1)). Of these, 74 had low burden (N + (1~2)) and 44 high burden (N + (≥ 3)). The remaining 156 patients had none (N0). AUC values across the 100 test repeats in predicting ALN status (N0/N + (≥ 1)) were 0.75 ± 0.05 (0.67~0.93, radiomics model), 0.68 ± 0.07 (0.53~0.85, clinical model), and 0.74 ± 0.05 (0.67~0.88, combined model). For metastatic burden prediction (N + (1~2)/N + (≥ 3)), AUC values were 0.65 ± 0.10 (0.50~0.88, radiomics model), 0.55 ± 0.10 (0.40~0.80, clinical model), and 0.64 ± 0.09 (0.50~0.90, combined model), with all the ranges spanning 0.5. In both cases, the radiomics model was significantly better than the clinical model (both p < 0.01) and comparable with the combined model (p = 0.56 and 0.64). CONCLUSIONS: Radiomics features of primary tumors could have potential in predicting ALN metastasis in CE-CBBCT imaging. CLINICAL RELEVANCE STATEMENT: The findings support potential clinical use of radiomics for predicting axillary lymph node metastasis in breast cancer patients and addressing the limited axilla coverage of cone-beam breast CT. KEY POINTS: • Contrast-enhanced cone-beam breast CT-based radiomics could have potential to predict N0 vs. N + (≥ 1) and, to a limited extent, N + (1~2) vs. N + (≥ 3) from primary tumor, and this could help address the limited axilla coverage, pending future verifications on larger cohorts. • The average AUC of radiomics and combined models was significantly higher than that of clinical models but showed no significant difference between themselves. • Radiomics features descriptive of tumor texture were found informative on axillary lymph node status, highlighting a higher heterogeneity for tumor with positive axillary lymph node.

Radiation dose reduction of 50% in dynamic myocardial CT perfusion with skipped beat acquisition: a retrospective study.

BACKGROUND: Dynamic myocardial computed tomography perfusion (CTP) is a novel imaging technique that increases the applicability of CT for cardiac imaging; however, the scanning requires a substantial radiation dose. PURPOSE: To investigate the feasibility of dose reduction in dynamic CTP by comparing all-heartbeat acquisitions to periodic skipping of heartbeats. MATERIAL AND METHODS: We retrieved imaging data of 38 dynamic CTP patients and created new datasets with every fourth, third or second beat (Skip1:4, Skip1:3, Skip1:2, respectively) removed. Seven observers evaluated the resulting images and perfusion maps for perfusion deficits. The mean blood flow (MBF) in each of the 16 myocardial segments was compared per skipped-beat level, normalized by the respective MBF for the full dose, and averaged across patients. The number of segments/cases whose MBF was <1.0 mL/g/min were counted. RESULTS: Out of 608 segments in 38 cases, the total additional number of false-negative (FN) segments over those present in the full-dose acquisitions and the number of additional false-positive cases were shown as acquisition (segment [%], case): Skip1:4: 7 (1.2%, 1); Skip1:3: 12 (2%, 3), and Skip1:2: 5 (0.8%, 2). The variability in quantitative MBF analysis in the repeated analysis for the reference condition resulted in 8 (1.3%) additional FN segments. The normalized results show a comparable MBF across all segments and patients, with relative mean MBFs as 1.02 ± 0.16, 1.03 ± 0.25, and 1.06 ± 0.30 for the Skip1:4, Skip1:3, and Skip1:2 protocols, respectively. CONCLUSION: Skipping every second beat acquisition during dynamic myocardial CTP appears feasible and may result in a radiation dose reduction of 50%. Diagnostic performance does not decrease after removing 50% of time points in dynamic sequence.

Optimizing the Pairs of Radiologists That Double Read Screening Mammograms.

Background Despite variation in performance characteristics among radiologists, the pairing of radiologists for the double reading of screening mammograms is performed randomly. It is unknown how to optimize pairing to improve screening performance. Purpose To investigate whether radiologist performance characteristics can be used to determine the optimal set of pairs of radiologists to double read screening mammograms for improved accuracy. Materials and Methods This retrospective study was performed with reading outcomes from breast cancer screening programs in Sweden (2008-2015), England (2012-2014), and Norway (2004-2018). Cancer detection rates (CDRs) and abnormal interpretation rates (AIRs) were calculated, with AIR defined as either reader flagging an examination as abnormal. Individual readers were divided into performance categories based on their high and low CDR and AIR. The performance of individuals determined the classification of pairs. Random pair performance, for which any type of pair was equally represented, was compared with the performance of specific pairing strategies, which consisted of pairs of readers who were either opposite or similar in AIR and/or CDR. Results Based on a minimum number of examinations per reader and per pair, the final study sample consisted of 3 592 414 examinations (Sweden, n = 965 263; England, n = 837 048; Norway, n = 1 790 103). The overall AIRs and CDRs for all specific pairing strategies (Sweden AIR range, 45.5-56.9 per 1000 examinations and CDR range, 3.1-3.6 per 1000; England AIR range, 68.2-70.5 per 1000 and CDR range, 8.9-9.4 per 1000; Norway AIR range, 81.6-88.1 per 1000 and CDR range, 6.1-6.8 per 1000) were not significantly different from the random pairing strategy (Sweden AIR, 54.1 per 1000 examinations and CDR, 3.3 per 1000; England AIR, 69.3 per 1000 and CDR, 9.1 per 1000; Norway AIR, 84.1 per 1000 and CDR, 6.3 per 1000). Conclusion Pairing a set of readers based on different pairing strategies did not show a significant difference in screening performance when compared with random pairing. © RSNA, 2023.

Dynamic myocardial CT perfusion imaging-state of the art.

In patients with suspected coronary artery disease (CAD), dynamic myocardial computed tomography perfusion (CTP) imaging combined with coronary CT angiography (CTA) has become a comprehensive diagnostic examination technique resulting in both anatomical and quantitative functional information on myocardial blood flow, and the presence and grading of stenosis. Recently, CTP imaging has been proven to have good diagnostic accuracy for detecting myocardial ischemia, comparable to stress magnetic resonance imaging and positron emission tomography perfusion, while being superior to single photon emission computed tomography. Dynamic CTP accompanied by coronary CTA can serve as a gatekeeper for invasive workup, as it reduces unnecessary diagnostic invasive coronary angiography. Dynamic CTP also has good prognostic value for the prediction of major adverse cardiovascular events. In this article, we will provide an overview of dynamic CTP, including the basics of coronary blood flow physiology, applications and technical aspects including protocols, image acquisition and reconstruction, future perspectives, and scientific challenges. KEY POINTS: • Stress dynamic myocardial CT perfusion combined with coronary CTA is a comprehensive diagnostic examination technique resulting in both anatomical and quantitative functional information. • Dynamic CTP imaging has good diagnostic accuracy for detecting myocardial ischemia comparable to stress MRI and PET perfusion. • Dynamic CTP accompanied by coronary CTA may serve as a gatekeeper for invasive workup and can guide treatment in obstructive coronary artery disease.

Abdominopelvic CT Image Quality: Evaluation of Thin (0.5-mm) Slices Using Deep Learning Reconstruction.

BACKGROUND. Because thick-section images (typically 3-5 mm) have low image noise, radiologists typically use them to perform clinical interpretation, although they may additionally refer to thin-section images (typically 0.5-0.625 mm) for problem solving. Deep learning reconstruction (DLR) can yield thin-section images with low noise. OBJECTIVE. The purpose of this study is to compare abdominopelvic CT image quality between thin-section DLR images and thin- and thick-section hybrid iterative reconstruction (HIR) images. METHODS. This retrospective study included 50 patients (31 men and 19 women; median age, 64 years) who underwent abdominopelvic CT between June 15, 2020, and July 29, 2020. Images were reconstructed at 0.5-mm section using DLR and at 0.5-mm and 3.0-mm sections using HIR. Five radiologists independently performed pairwise comparisons (0.5-mm DLR and either 0.5-mm or 3.0-mm HIR) and recorded the preferred image for subjective image quality measures (scale, -2 to 2). The pooled scores of readers were compared with a score of 0 (denoting no preference). Image noise was quantified using the SD of ROIs on regions of homogeneous liver. RESULTS. For comparison of 0.5-mm DLR images and 0.5-mm HIR images, the median pooled score was 2 (indicating a definite preference for DLR) for noise and overall image quality and 1 (denoting a slight preference for DLR) for sharpness and natural appearance. For comparison of 0.5-mm DLR and 3.0-mm HIR, the median pooled score was 1 for the four previously mentioned measures. These assessments were all significantly different (p < .001) from 0. For artifacts, the median pooled score for both comparisons was 0, which was not significant for comparison with 3.0-mm HIR (p = .03) but was significant for comparison with 0.5-mm HIR (p < .001) due to imbalance in scores of 1 (n = 28) and -1 (slight preference for HIR, n = 1). Noise for 0.5-mm DLR was lower by mean differences of 12.8 HU compared with 0.5-mm HIR and 4.4 HU compared with 3.0-mm HIR (both p < .001). CONCLUSION. Thin-section DLR improves subjective image quality and reduces image noise compared with currently used thin- and thick-section HIR, without causing additional artifacts. CLINICAL IMPACT. Although further diagnostic performance studies are warranted, the findings suggest the possibility of replacing current use of both thin- and thick-section HIR with the use of thin-section DLR only during clinical interpretations.

Multireader image quality evaluation of dynamic myocardial computed tomography perfusion imaging with a novel four-dimensional noise reduction filter.

BACKGROUND: Dynamic myocardial computed tomography perfusion (CTP) is a novel technique able to depict cardiac ischemia. PURPOSE: To evaluate the impact of a four-dimensional noise reduction filter (similarity filter [4D-SF]) on image quality in dynamic CTP imaging, allowing for substantial radiation dose reduction. MATERIAL AND METHODS: Dynamic CTP datasets of 30 patients (16 women) with suspected coronary artery disease, acquired with a 320-slice CT system, were retrieved, reconstructed with the deep learning-based algorithm of the system (DLR), and filtered with the 4D-SF. For each case, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in six regions of interest (33-38mm2) were calculated before and after filtering, in four-chamber and short-axis views, and t-tested. Furthermore, six radiologists of different expertise evaluated subjective image preference by answering five visual grading analysis-type questions (regarding acceptable level of noise, absence of artifacts, natural appearance, cardiac contour sharpness, diagnostic acceptability) using a 5-point scale. The results were analyzed using visual grade characteristics (VGC) and intraclass correlation coefficient (ICC). RESULTS: Mean SNR in four-chamber view (unfiltered vs. filtered) were: septum=4.1 ± 2.1 versus 7.6 ± 5.6; lateral wall=4.5 ± 2.0 versus 8.0 ± 4.9; CNRseptum=16.6 ± 8.9 versus 31.7 ± 28; lateral wall=16.2 ± 8.9 versus 31.3 ± 28.9. Similar results were obtained in short-axis view. The perceived filtered image quality indicated decreased noise (VGCAUC=0.96) and artifacts (0.65), improved natural appearance (0.59), cardiac contour sharpness (0.74), and diagnostic acceptability (0.78). The inter-observer variability was excellent (ICC=0.79). All results were statistically significant (P < 0.05). CONCLUSION: Similarity filtering after DLR improves image quality, possibly enabling dose reduction in dynamic CTP imaging in patient with suspected chronic coronary syndrome.

Artificial intelligence for breast cancer detection in mammography and digital breast tomosynthesis: State of the art.

Screening for breast cancer with mammography has been introduced in various countries over the last 30 years, initially using analog screen-film-based systems and, over the last 20 years, transitioning to the use of fully digital systems. With the introduction of digitization, the computer interpretation of images has been a subject of intense interest, resulting in the introduction of computer-aided detection (CADe) and diagnosis (CADx) algorithms in the early 2000's. Although they were introduced with high expectations, the potential improvement in the clinical realm failed to materialize, mostly due to the high number of false positive marks per analyzed image. In the last five years, the artificial intelligence (AI) revolution in computing, driven mostly by deep learning and convolutional neural networks, has also pervaded the field of automated breast cancer detection in digital mammography and digital breast tomosynthesis. Research in this area first involved comparison of its capabilities to that of conventional CADe/CADx methods, which quickly demonstrated the potential of this new technology. In the last couple of years, more mature and some commercial products have been developed, and studies of their performance compared to that of experienced breast radiologists are showing that these algorithms are on par with human-performance levels in retrospective data sets. Although additional studies, especially prospective evaluations performed in the real screening environment, are needed, it is becoming clear that AI will have an important role in the future breast cancer screening realm. Exactly how this new player will shape this field remains to be determined, but recent studies are already evaluating different options for implementation of this technology. The aim of this review is to provide an overview of the basic concepts and developments in the field AI for breast cancer detection in digital mammography and digital breast tomosynthesis. The pitfalls of conventional methods, and how these are, for the most part, avoided by this new technology, will be discussed. Importantly, studies that have evaluated the current capabilities of AI and proposals for how these capabilities should be leveraged in the clinical realm will be reviewed, while the questions that need to be answered before this vision becomes a reality are posed.

Interval Cancer Detection Using a Neural Network and Breast Density in Women with Negative Screening Mammograms.

Background Inclusion of mammographic breast density (BD) in breast cancer risk models improves accuracy, but accuracy remains modest. Interval cancer (IC) risk prediction may be improved by combining assessments of BD and an artificial intelligence (AI) cancer detection system. Purpose To evaluate the performance of a neural network (NN)-based model that combines the assessments of BD and an AI system in the prediction of risk of developing IC among women with negative screening mammography results. Materials and Methods This retrospective nested case-control study performed with screening examinations included women who developed IC and women with normal follow-up findings (from January 2011 to January 2015). An AI cancer detection system analyzed all studies yielding a score of 1-10, representing increasing likelihood of malignancy. BD was automatically computed using publicly available software. An NN model was trained by combining the AI score and BD using 10-fold cross-validation. Bootstrap analysis was used to calculate the area under the receiver operating characteristic curve (AUC), sensitivity at 90% specificity, and 95% CIs of the AI, BD, and NN models. Results A total of 2222 women with IC and 4661 women in the control group were included (mean age, 61 years; age range, 49-76 years). AUC of the NN model was 0.79 (95% CI: 0.77,0.81), which was higher than AUC of the AI cancer detection system or BD alone (AUC, 0.73 [95% CI: 0.71, 0.76] and 0.69 [95% CI: 0.67, 0.71], respectively; P < .001 for both). At 90% specificity, the NN model had a sensitivity of 50.9% (339 of 666 women; 95% CI: 45.2, 56.3) for prediction of IC, which was higher than that of the AI system (37.5%; 250 of 666 women; 95% CI: 33.0, 43.7; P < .001) or BD percentage alone (22.4%; 149 of 666 women; 95% CI: 17.9, 28.5; P < .001). Conclusion The combined assessment of an artificial intelligence detection system and breast density measurements enabled identification of a larger proportion of women who would develop interval cancer compared with either method alone. Published under a CC BY 4.0 license.

Evaluation of reader performance during interpretation of breast cancer screening: the Recall and detection Of breast Cancer in Screening (ROCS) trial study design.

The magnitude of the tradeoff between recall rate (RR) and cancer detection rate (CDR) in breast-cancer screening is not clear, and it is expected to depend on target population and screening program characteristics. Multi-reader multi-case research studies, which may be used to estimate this tradeoff, rely on enriched datasets with artificially high prevalence rates, which may bias the results. Furthermore, readers participating in research studies are subject to "laboratory" effects, which can alter their performance relative to actual practice. The Recall and detection Of breast Cancer in Screening (ROCS) trial uses a novel data acquisition system that minimizes these limitations while obtaining an estimate of the RR-CDR curve during actual practice in the Dutch National Breast Cancer Screening Program. ROCS involves collection of at least 40,000 probability-of-malignancy ratings from at least 20 radiologists during interpretation of approximately 2,000 digital mammography screening cases each. With the use of custom-built software on a tablet, and a webcam, this data was obtained in the usual reading environment with minimal workflow disruption and without electronic access to the review workstation software. Comparison of the results to short- and medium-term follow-up allows for estimation of the RR-CDR and receiver operating characteristics curves, respectively. The anticipated result of the study is that performance-based evidence from practice will be available to determine the optimal operating point for breast-cancer screening. In addition, this data will be useful as a benchmark when evaluating the impact of potential new screening technologies, such as digital breast tomosynthesis or artificial intelligence. KEY POINTS: • The ROCS trial aims to estimate the recall rate-cancer detection rate curve during actual screening practice in the Dutch National Breast Cancer Screening Program. • The study design is aimed at avoiding the influence of the "laboratory effect" in usual observer performance studies. • The use of a tablet and a webcam allows for the acquisition of probability of malignancy ratings without access to the review workstation software.

Impact of Artificial Intelligence Decision Support Using Deep Learning on Breast Cancer Screening Interpretation with Single-View Wide-Angle Digital Breast Tomosynthesis.

Background The high volume of data in digital breast tomosynthesis (DBT) and the lack of agreement on how to best implement it in screening programs makes its use challenging. Purpose To compare radiologist performance when reading single-view wide-angle DBT images with and without an artificial intelligence (AI) system for decision and navigation support. Materials and Methods A retrospective observer study was performed with bilateral mediolateral oblique examinations and corresponding synthetic two-dimensional images acquired between June 2016 and February 2018 with a wide-angle DBT system. Fourteen breast screening radiologists interpreted 190 DBT examinations (90 normal, 26 with benign findings, and 74 with malignant findings), with the reference standard being verified by using histopathologic analysis or at least 1 year of follow-up. Reading was performed in two sessions, separated by at least 4 weeks, with a random mix of examinations being read with and without AI decision and navigation support. Forced Breast Imaging Reporting and Data System (categories 1-5) and level of suspicion (1-100) scores were given per breast by each reader. The area under the receiver operating characteristic curve (AUC) and the sensitivity and specificity were compared between conditions by using the public-domain iMRMC software. The average reading times were compared by using the Wilcoxon signed rank test. Results The 190 women had a median age of 54 years (range, 48-63 years). The examination-based reader-averaged AUC was higher when interpreting results with AI support than when reading unaided (0.88 [95% CI: 0.84, 0.92] vs 0.85 [95% CI: 0.80, 0.89], respectively; P = .01). The average sensitivity increased with AI support (64 of 74, 86% [95% CI: 80%, 92%] vs 60 of 74, 81% [95% CI: 74%, 88%]; P = .006), whereas no differences in the specificity (85 of 116, 73.3% [95% CI: 65%, 81%] vs 83 of 116, 71.6% [95% CI: 65%, 78%]; P = .48) or reading time (48 seconds vs 45 seconds; P = .35) were detected. Conclusion Using a single-view digital breast tomosynthesis (DBT) and artificial intelligence setup could allow for a more effective screening program with higher performance, especially in terms of an increase in cancers detected, than using single-view DBT alone. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Chan and Helvie in this issue.

How does image quality affect radiologists' perceived ability for image interpretation and lesion detection in digital mammography?

OBJECTIVES: To study how radiologists' perceived ability to interpret digital mammography (DM) images is affected by decreases in image quality. METHODS: One view from 45 DM cases (including 30 cancers) was degraded to six levels each of two acquisition-related issues (lower spatial resolution and increased quantum noise) and three post-processing-related issues (lower and higher contrast and increased correlated noise) seen during clinical evaluation of DM systems. The images were shown to fifteen breast screening radiologists from five countries. Aware of lesion location, the radiologists selected the most-degraded mammogram (indexed from 1 (reference) to 7 (most degraded)) they still felt was acceptable for interpretation. The median selected index, per degradation type, was calculated separately for calcification and soft tissue (including normal) cases. Using the two-sided, non-parametric Mann-Whitney test, the median indices for each case and degradation type were compared. RESULTS: Radiologists were not tolerant to increases (medians: 1.5 (calcifications) and 2 (soft tissue)) or decreases (median: 2, for both types) in contrast, but were more tolerant to correlated noise (median: 3, for both types). Increases in quantum noise were tolerated more for calcifications than for soft tissue cases (medians: 3 vs. 4, p = 0.02). Spatial resolution losses were considered less acceptable for calcification detection than for soft tissue cases (medians: 3.5 vs. 5, p = 0.001). CONCLUSIONS: Perceived ability of radiologists for image interpretation in DM was affected not only by image acquisition-related issues but also by image post-processing issues, and some of those issues affected calcification cases more than soft tissue cases. KEY POINTS: • Lower spatial resolution and increased quantum noise affected the radiologists' perceived ability to interpret calcification cases more than soft tissue lesion or normal cases. • Post-acquisition image processing-related effects, not only image acquisition-related effects, also impact the perceived ability of radiologists to interpret images and detect lesions. • In addition to current practices, post-acquisition image processing-related effects need to also be considered during the testing and evaluation of digital mammography systems.

Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study.

OBJECTIVES: Digital breast tomosynthesis (DBT) increases sensitivity of mammography and is increasingly implemented in breast cancer screening. However, the large volume of images increases the risk of reading errors and reading time. This study aims to investigate whether the accuracy of breast radiologists reading wide-angle DBT increases with the aid of an artificial intelligence (AI) support system. Also, the impact on reading time was assessed and the stand-alone performance of the AI system in the detection of malignancies was compared to the average radiologist. METHODS: A multi-reader multi-case study was performed with 240 bilateral DBT exams (71 breasts with cancer lesions, 70 breasts with benign findings, 339 normal breasts). Exams were interpreted by 18 radiologists, with and without AI support, providing cancer suspicion scores per breast. Using AI support, radiologists were shown examination-based and region-based cancer likelihood scores. Area under the receiver operating characteristic curve (AUC) and reading time per exam were compared between reading conditions using mixed-models analysis of variance. RESULTS: On average, the AUC was higher using AI support (0.863 vs 0.833; p = 0.0025). Using AI support, reading time per DBT exam was reduced (p < 0.001) from 41 (95% CI = 39-42 s) to 36 s (95% CI = 35- 37 s). The AUC of the stand-alone AI system was non-inferior to the AUC of the average radiologist (+0.007, p = 0.8115). CONCLUSIONS: Radiologists improved their cancer detection and reduced reading time when evaluating DBT examinations using an AI reading support system. KEY POINTS: • Radiologists improved their cancer detection accuracy in digital breast tomosynthesis (DBT) when using an AI system for support, while simultaneously reducing reading time. • The stand-alone breast cancer detection performance of an AI system is non-inferior to the average performance of radiologists for reading digital breast tomosynthesis exams. • The use of an AI support system could make advanced and more reliable imaging techniques more accessible and could allow for more cost-effective breast screening programs with DBT.

Deep learning-based reconstruction may improve non-contrast cerebral CT imaging compared to other current reconstruction algorithms.

OBJECTIVES: To evaluate image quality and reconstruction times of a commercial deep learning reconstruction algorithm (DLR) compared to hybrid-iterative reconstruction (Hybrid-IR) and model-based iterative reconstruction (MBIR) algorithms for cerebral non-contrast CT (NCCT). METHODS: Cerebral NCCT acquisitions of 50 consecutive patients were reconstructed using DLR, Hybrid-IR and MBIR with a clinical CT system. Image quality, in terms of six subjective characteristics (noise, sharpness, grey-white matter differentiation, artefacts, natural appearance and overall image quality), was scored by five observers. As objective metrics of image quality, the noise magnitude and signal-difference-to-noise ratio (SDNR) of the grey and white matter were calculated. Mean values for the image quality characteristics scored by the observers were estimated using a general linear model to account for multiple readers. The estimated means for the reconstruction methods were pairwise compared. Calculated measures were compared using paired t tests. RESULTS: For all image quality characteristics, DLR images were scored significantly higher than MBIR images. Compared to Hybrid-IR, perceived noise and grey-white matter differentiation were better with DLR, while no difference was detected for other image quality characteristics. Noise magnitude was lower for DLR compared to Hybrid-IR and MBIR (5.6, 6.4 and 6.2, respectively) and SDNR higher (2.4, 1.9 and 2.0, respectively). Reconstruction times were 27 s, 44 s and 176 s for Hybrid-IR, DLR and MBIR respectively. CONCLUSIONS: With a slight increase in reconstruction time, DLR results in lower noise and improved tissue differentiation compared to Hybrid-IR. Image quality of MBIR is significantly lower compared to DLR with much longer reconstruction times. KEY POINTS: • Deep learning reconstruction of cerebral non-contrast CT results in lower noise and improved tissue differentiation compared to hybrid-iterative reconstruction. • Deep learning reconstruction of cerebral non-contrast CT results in better image quality in all aspects evaluated compared to model-based iterative reconstruction. • Deep learning reconstruction only needs a slight increase in reconstruction time compared to hybrid-iterative reconstruction, while model-based iterative reconstruction requires considerably longer processing time.

Correction to: Physical evaluation of an ultra-high-resolution CT scanner.

The original version of this article, published on 10 February 2019, unfortunately contained a mistake. The axes of the graphs in Fig. 3 are incorrect. The correct figure is given below. Therefore, the last two sentences in "Results," section "Noise," should read: "The peak frequency of the HR and SHR was 0.21 lp/mm. For the NR mode and the MDCT, the peak frequencies were 0.17 lp/mm and 0.21 lp/mm, respectively."

Physical evaluation of an ultra-high-resolution CT scanner.

OBJECTIVES: To evaluate the technical performance of an ultra-high-resolution CT (UHRCT) system. METHODS: The physico-technical capabilities of a novel commercial UHRCT system were assessed and compared with those of a current-generation multi-detector (MDCT) system. The super-high-resolution (SHR) mode of the system uses 0.25 mm (at isocentre) detector elements (dels) in the in-plane and longitudinal directions, while the high-resolution (HR) mode bins two dels in the longitudinal direction. The normal-resolution (NR) mode bins dels 2 × 2, resulting in a del-size equivalent to that of the MDCT system. In general, standard procedures and phantoms were used to perform these assessments. RESULTS: The UHRCT MTF (10% MTF 4.1 lp/mm) is twice as high as that of the MDCT (10% MTF 1.9 lp/mm), which is comparable to the MTF in the NR mode (10% MTF 1.7 lp/mm). The width of the slice sensitivity profile in the SHR mode (FWHM 0.45 mm) is about 60% of that of the MDCT (FWHM 0.77 mm). Uniformity and CT numbers are within the expected range. Noise in the high-resolution modes has a higher magnitude and higher frequency components compared with MDCT. Low-contrast visibility is lower for the NR, HR and SHR modes compared with MDCT, but about a 14%, for NR, and 23%, for HR and SHR, dose increase gives the same results. CONCLUSIONS: HR and SHR mode scanning results in double the spatial resolution, with about a 23% increase in dose required to achieve the same low-contrast detectability. KEY POINTS: • Resolution on UHRCT is up to twice as high as for the tested MDCT. • With abdominal settings, UHRCT needs higher dose for the same low-contrast detectability as MDCT, but dose is still below achievable levels as defined by current diagnostic reference levels. • The UHRCT system used in normal-resolution mode yields comparable resolution and noise characteristics as the MDCT system.

Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System.

Purpose To compare breast cancer detection performance of radiologists reading mammographic examinations unaided versus supported by an artificial intelligence (AI) system. Materials and Methods An enriched retrospective, fully crossed, multireader, multicase, HIPAA-compliant study was performed. Screening digital mammographic examinations from 240 women (median age, 62 years; range, 39-89 years) performed between 2013 and 2017 were included. The 240 examinations (100 showing cancers, 40 leading to false-positive recalls, 100 normal) were interpreted by 14 Mammography Quality Standards Act-qualified radiologists, once with and once without AI support. The readers provided a Breast Imaging Reporting and Data System score and probability of malignancy. AI support provided radiologists with interactive decision support (clicking on a breast region yields a local cancer likelihood score), traditional lesion markers for computer-detected abnormalities, and an examination-based cancer likelihood score. The area under the receiver operating characteristic curve (AUC), specificity and sensitivity, and reading time were compared between conditions by using mixed-models analysis dof variance and generalized linear models for multiple repeated measurements. Results On average, the AUC was higher with AI support than with unaided reading (0.89 vs 0.87, respectively; P = .002). Sensitivity increased with AI support (86% [86 of 100] vs 83% [83 of 100]; P = .046), whereas specificity trended toward improvement (79% [111 of 140]) vs 77% [108 of 140]; P = .06). Reading time per case was similar (unaided, 146 seconds; supported by AI, 149 seconds; P = .15). The AUC with the AI system alone was similar to the average AUC of the radiologists (0.89 vs 0.87). Conclusion Radiologists improved their cancer detection at mammography when using an artificial intelligence system for support, without requiring additional reading time. Published under a CC BY 4.0 license. See also the editorial by Bahl in this issue.

Iodine Maps from Subtraction CT or Dual-Energy CT to Detect Pulmonary Emboli with CT Angiography: A Multiple-Observer Study.

Background Dual-energy CT iodine maps are used to detect pulmonary embolism (PE) with CT angiography but require dedicated hardware. Subtraction CT, a software-only solution, results in iodine maps with high contrast-to-noise ratios. Purpose To compare the use of subtraction CT versus dual-energy CT iodine maps to CT angiography for PE detection. Materials and Methods In this prospective study ( https://clinicaltrials.gov , NCT02890706), 274 participants suspected of having PE underwent precontrast CT followed by contrast material-enhanced dual-energy CT angiography between July 2016 and April 2017. Iodine maps from dual-energy CT were derived. Subtraction maps (contrast-enhanced CT minus precontrast CT) were calculated after motion correction. Truth was established by expert consensus. A total of 75 randomly selected participants with and without PE (1:1 ratio) were evaluated by three radiologists and six radiology residents (blinded to final diagnosis) for the presence of PE using three types of CT: CT angiography alone, dual-energy CT, and subtraction CT. The partial area under the receiver operating characteristic curve (AUC) for the clinically relevant specificity region (maximum partial AUC, 0.11) was compared by using multireader multicase variance. A P value less than or equal to .025 was considered indicative of a significant difference due to multiple comparisons. Results There were 35 men and 40 women in the reader study (mean age, 63 years ± 12 [standard deviation]). The pooled sensitivities were not different (P ≥ .31 among techniques) (95% confidence intervals [CIs]: 67%, 89% for CT angiography; 72%, 91% for dual-energy CT; 70%, 91% for subtraction CT). However, pooled specificity was higher for subtraction CT (95% CI: 100%, 100%) than for CT angiography (95% CI: 89%, 97%) or dual-energy CT (95% CI: 89%, 98%) (P < .001). Partial AUCs for the average observer improved equally when adding iodine maps (subtraction CT [0.093] vs CT angiography [0.088], P = .03; dual-energy CT [0.094] vs CT angiography, P = .01; dual-energy CT vs subtraction CT, P = .68). Average reading times were equivalent (range, 97-101 seconds; P ≥ .41) among techniques. Conclusion Subtraction CT iodine maps had greater specificity than CT angiography alone in pulmonary embolism detection. Subtraction CT had comparable diagnostic performance to that of dual-energy CT, without the need for dedicated hardware. © RSNA, 2019 Online supplemental material is available for this article.

Imaging of pulmonary perfusion using subtraction CT angiography is feasible in clinical practice.

Subtraction computed tomography (SCT) is a technique that uses software-based motion correction between an unenhanced and an enhanced CT scan for obtaining the iodine distribution in the pulmonary parenchyma. This technique has been implemented in clinical practice for the evaluation of lung perfusion in CT pulmonary angiography (CTPA) in patients with suspicion of acute and chronic pulmonary embolism, with acceptable radiation dose. This paper discusses the technical principles, clinical interpretation, benefits and limitations of arterial subtraction CTPA. KEY POINTS: • SCT uses motion correction and image subtraction between an unenhanced and an enhanced CT scan to obtain iodine distribution in the pulmonary parenchyma. • SCT could have an added value in detection of pulmonary embolism. • SCT requires only software implementation, making it potentially more widely available for patient care than dual-energy CT.

Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.

PURPOSE: To study the feasibility of automatically identifying normal digital mammography (DM) exams with artificial intelligence (AI) to reduce the breast cancer screening reading workload. METHODS AND MATERIALS: A total of 2652 DM exams (653 cancer) and interpretations by 101 radiologists were gathered from nine previously performed multi-reader multi-case receiver operating characteristic (MRMC ROC) studies. An AI system was used to obtain a score between 1 and 10 for each exam, representing the likelihood of cancer present. Using all AI scores between 1 and 9 as possible thresholds, the exams were divided into groups of low- and high likelihood of cancer present. It was assumed that, under the pre-selection scenario, only the high-likelihood group would be read by radiologists, while all low-likelihood exams would be reported as normal. The area under the reader-averaged ROC curve (AUC) was calculated for the original evaluations and for the pre-selection scenarios and compared using a non-inferiority hypothesis. RESULTS: Setting the low/high-likelihood threshold at an AI score of 5 (high likelihood > 5) results in a trade-off of approximately halving (- 47%) the workload to be read by radiologists while excluding 7% of true-positive exams. Using an AI score of 2 as threshold yields a workload reduction of 17% while only excluding 1% of true-positive exams. Pre-selection did not change the average AUC of radiologists (inferior 95% CI > - 0.05) for any threshold except at the extreme AI score of 9. CONCLUSION: It is possible to automatically pre-select exams using AI to significantly reduce the breast cancer screening reading workload. KEY POINTS: • There is potential to use artificial intelligence to automatically reduce the breast cancer screening reading workload by excluding exams with a low likelihood of cancer. • The exclusion of exams with the lowest likelihood of cancer in screening might not change radiologists' breast cancer detection performance. • When excluding exams with the lowest likelihood of cancer, the decrease in true-positive recalls would be balanced by a simultaneous reduction in false-positive recalls.

Image Quality of Iodine Maps for Pulmonary Embolism: A Comparison of Subtraction CT and Dual-Energy CT.

OBJECTIVE. The objective of this study was to compare the image quality of iodine maps derived from subtraction CT and from dual-energy CT (DECT) in patients with suspected pulmonary embolism (PE). SUBJECTS AND METHODS. In this prospective study conducted between July 2016 and April 2017, consecutive patients with suspected PE underwent unenhanced CT at 100 kV and dual-energy pulmonary CT angiography at 100 and 140 kV on a dual-source scanner. The scanner was set to generate subtraction and DECT iodine maps at similar radiation doses. In 55 patients (30 women, 25 men; mean age ± SD, 63.4 ± 11.9 years old), various subjective image quality criteria including diagnostic acceptability were rated on a 5-point scale by four radiologists and a radiology resident. In 29 patients (17 women, 12 men; mean age, 62.4 ± 11.7 years old) with confirmed perfusion defects, the signal-difference-to-noise ratio (SDNR) between perfusion defects and adjacent normally perfused parenchyma was measured in corresponding ROIs on subtraction and DECT iodine maps. McNemar and Wilcoxon signed-rank tests were used for statistical comparisons. RESULTS. Diagnostic acceptability was rated excellent or good in a mean of 67% (range, 31-80%) of subtraction CT studies and 36% (5-69%) of DECT studies (p < 0.05 for four of the five radiologists), mainly because of fewer artifacts on subtraction CT. Mean SDNR was marginally higher for subtraction CT than for DECT (18.6 vs 17.1, p = 0.06) and was significantly higher in the upper lobes (21.8 vs 17.9, p < 0.05). CONCLUSION. Radiologist-judged image quality of pulmonary iodine maps was higher for subtraction CT than for DECT with similar to higher SDNR. Subtraction CT is a software-only solution, so it may be an attractive alternative to DECT for depicting perfusion defects.