Lung-RADS Version 1.1: Challenges and a Look Ahead, From the AJR Special Series on Radiology Reporting and Data Systems

Baseline Cell-Free DNA Can Predict Malignancy of Nodules Observed in the ITALUNG Screening Trial

The role of total plasma cell-free DNA (cfDNA) in lung cancer (LC) screening with low-dose computed tomography (LDCT) is uncertain. We hypothesized that cfDNA could support differentiation between malignant and benign nodules observed in LDCT. The baseline cfDNA was measured in 137 subjects of the ITALUNG trial, including 29 subjects with screen-detected LC (17 prevalent and 12 incident) and 108 subjects with benign nodules. The predictive capability of baseline cfDNA to differentiate malignant and benign nodules was compared to that of Lung-RADS classification and Brock score at initial LDCT (iLDCT). Subjects with prevalent LC showed both well-discriminating radiological characteristics of the malignant nodule (16 of 17 were classified as Lung-RADS 4) and markedly increased cfDNA (mean 18.8 ng/mL). The mean diameters and Brock scores of malignant nodules at iLDCT in subjects who were diagnosed with incident LC were not different from those of benign nodules. However, 75% (9/12) of subjects with incident LC showed a baseline cfDNA ≥ 3.15 ng/mL, compared to 34% (37/108) of subjects with benign nodules (p = 0.006). Moreover, baseline cfDNA was correlated (p = 0.001) with tumor growth, measured with volume doubling time. In conclusion, increased baseline cfDNA may help to differentiate subjects with malignant and benign nodules at LDCT.

Axial Spondyloarthritis: Does Magnetic Resonance Imaging Classification Improve Report Interpretation

OBJECTIVE

The interpretation of magnetic resonance imaging (MRI) reports is crucial for the diagnosis of axial spondyloarthritis, but the subjective nature of narrative reports can lead to varying interpretations. This study presents a validation of a novel MRI reporting system for the sacroiliac joint in clinical practice.

METHODS

A historical review was conducted on 130 consecutive patients referred by 2 rheumatologists for initial MRI assessment of possible axial spondyloarthritis. The original MRI reports were interpreted by the rheumatologists and the radiologist who originally read the images and then categorized according to the novel system. Two musculoskeletal radiologists then reinterpreted the original MRI scans using the new system, and the resulting reports were interpreted and categorized by the same rheumatologists. The quality of the new framework was assessed by comparing the interpretations of both reports.

RESULTS

Ninety-two patients met the study criteria. The rheumatologists disagreed on the categorization of the original MRI reports in 12% of cases. The rheumatologists and original radiologists disagreed on the categorization of the initial report in 23.4% of cases. In contrast, there was 100% agreement between the rheumatologists and radiologists on the categorization of the new MRI report.

CONCLUSION

The new MRI categorization system significantly improved the agreement between the clinician and radiologist in report interpretation. The system provided a standard vocabulary for reporting, reduced variability in report interpretation, and may therefore improve clinical decision-making.

Discrimination of invasive lung adenocarcinoma from Lung-RADS category 2 nonsolid nodules through visual assessment: a retrospective study

OBJECTIVES

Invasive adenocarcinomas (IADs) have been identified among nonsolid nodules (NSNs) assigned as Lung Imaging Reporting and Data System (Lung-RADS) category 2. This study used visual assessment for differentiating IADs from noninvasive lesions (NILs) in this category.

METHODS

This retrospective study included 222 patients with 242 NSNs, which were resected after preoperative computed tomography (CT)-guided dye localization. Visual assessment was performed by using the lung and bone window (BW) settings to classify NSNs into BW-visible (BWV) and BW-invisible (BWI) NSNs. In addition, nodule size, shape, border, CT attenuation, and location were evaluated and correlated with histopathological results. Logistic regression was performed for multivariate analysis. A p value of < 0.05 was considered statistically significant.

RESULTS

A total of 242 NSNs (mean diameter, 7.6 ± 2.8 mm), including 166 (68.6%) BWV and 76 (31.4%) BWI NSNs, were included. IADs accounted for 31% (75) of the nodules. Only 4 (5.3%) IADs were identified in the BWI group and belonged to the lepidic-predominant (n = 3) and acinar-predominant (n = 1) subtypes. In univariate analysis for differentiating IADs from NILs, the nodule size, shape, CT attenuation, and visual classification exhibited statistical significance. Nodule size and visual classification were the significant predictors for IAD in multivariate analysis with logistic regression (p < 0.05). The sensitivity, specificity, positive predictive value, and negative predictive value of visual classification in IAD prediction were 94.7%, 43.1%, 42.8%, and 94.7%, respectively.

CONCLUSIONS

The window-based visual classification of NSNs is a simple and objective method to discriminate IADs from NILs.

CLINICAL RELEVANCE STATEMENT

The present study shows that using the bone window to classify nonsolid nodules helps discriminate invasive adenocarcinoma from noninvasive lesions.

KEY POINTS

• Evidence has shown the presence of lung adenocarcinoma in Lung-RADS category 2 nonsolid nodules. • Nonsolid nodules are classified into the bone window-visible and the bone window-invisible nonsolid nodules, and this classification differentiates invasive adenocarcinoma from noninvasive lesions. • The Lung-RADS category 2 nonsolid nodules are unlikely invasive adenocarcinoma if they show nonvisualization in the bone window.

Assessment of Lung Nodule Detection and Lung CT Screening Reporting and Data System Classification Using Zero Echo Time Pulmonary MRI

BACKGROUND

The detection rate of lung nodules has increased considerably with CT as the primary method of examination, and the repeated CT examinations at 3 months, 6 months or annually, based on nodule characteristics, have increased the radiation exposure of patients. So, it is urgent to explore a radiation-free MRI examination method that can effectively address the challenges posed by low proton density and magnetic field inhomogeneities.

PURPOSE

To evaluate the potential of zero echo time (ZTE) MRI in lung nodule detection and lung CT screening reporting and data system (lung-RADS) classification, and to explore the value of ZTE-MRI in the assessment of lung nodules.

STUDY TYPE

Prospective.

POPULATION

54 patients, including 21 men and 33 women.

FIELD STRENGTH/SEQUENCE

Chest CT using a 16-slice scanner and ZTE-MRI at 3.0T based on fast gradient echo.

ASSESSMENT

Nodule type (ground-glass nodules, part-solid nodules, and solid nodules), lung-RADS classification, and nodule diameter (manual measurement) on CT and ZTE-MRI images were recorded.

STATISTICAL TESTS

The percent of concordant cases, Kappa value, intraclass correlation coefficient (ICC), Wilcoxon signed-rank test, Spearman's correlation, and Bland-Altman. The p-value <0.05 is considered significant.

RESULTS

A total of 54 patients (age, 54.8 ± 11.9 years; 21 men) with 63 nodules were enrolled. Compared with CT, the total nodule detection rate of ZTE-MRI was 85.7%. The intermodality agreement of ZTE-MRI and CT lung nodules type evaluation was substantial (Kappa = 0.761), and the intermodality agreement of ZTE-MRI and CT lung-RADS classification was moderate (Kappa = 0.592). The diameter measurements between ZTE-MRI and CT showed no significant difference and demonstrated a high degree of interobserver (ICC = 0.997-0.999) and intermodality (ICC = 0.956-0.985) agreements.

DATA CONCLUSION

The measurement of nodule diameter by pulmonary ZTE-MRI is similar to that by CT, but the ability of lung-RADS to classify nodes from MRI images still requires further research.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 2.

An alternating-intervention pilot trial on the impact of an informational handout on patient-reported outcomes and follow-up after lung cancer screening

INTRODUCTION

Lung cancer screening (LCS) can reduce lung cancer mortality; however, poor understanding of results may impact patient experience and follow-up. We sought to determine whether an informational handout accompanying LCS results can improve patient-reported outcomes and adherence to follow-up.

STUDY DESIGN

This was a prospective alternating intervention pilot trial of a handout to accompany LCS results delivery.

SETTING/PARTICIPANTS

Patients undergoing LCS in a multisite program over a 6-month period received a mailing containing either: 1) a standardized form letter of LCS results (control) or 2) the LCS results letter and the handout (intervention).

INTERVENTION

A two-sided informational handout on commonly asked questions after LCS created through iterative mixed-methods evaluation with both LCS patients and providers.

OUTCOME MEASURES

The primary outcomes of 1)patient understanding of LCS results, 2)correct identification of next steps in screening, and 3)patient distress were measured through survey. Adherence to recommended follow-up after LCS was determined through chart review. Outcomes were compared between the intervention and control group using generalized estimating equations.

RESULTS

389 patients were eligible and enrolled with survey responses from 230 participants (59% response rate). We found no differences in understanding of results, identification of next steps in follow-up or distress but did find higher levels of knowledge and understanding on questions assessing individual components of LCS in the intervention group. Follow-up adherence was overall similar between the two arms, though was higher in the intervention group among those with positive findings (p = 0.007).

CONCLUSIONS

There were no differences in self-reported outcomes between the groups or overall follow-up adherence. Those receiving the intervention did report greater understanding and knowledge of key LCS components, and those with positive results had a higher rate of follow-up. This may represent a feasible component of a multi-level intervention to address knowledge and follow-up for LCS.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05265897.

ACR Lung-RADS v2022: Assessment Categories and Management Recommendations

The ACR created the Lung CT Screening Reporting and Data System (Lung-RADS) in 2014 to standardize the reporting and management of screen-detected pulmonary nodules. Lung-RADS was updated to version 1.1 in 2019 and revised size thresholds for nonsolid nodules, added classification criteria for perifissural nodules, and allowed for short-interval follow-up of rapidly enlarging nodules that may be infectious in etiology. Lung-RADS v2022, released in November 2022, provides several updates including guidance on the classification and management of atypical pulmonary cysts, juxtapleural nodules, airway-centered nodules, and potentially infectious findings. This new release also provides clarification for determining nodule growth and introduces stepped management for nodules that are stable or decreasing in size. This article summarizes the current evidence and expert consensus supporting Lung-RADS v2022.

ACR Lung-RADS v2022: Assessment Categories and Management Recommendations

The American College of Radiology created the Lung CT Screening Reporting and Data System (Lung-RADS) in 2014 to standardize the reporting and management of screen-detected pulmonary nodules. Lung-RADS was updated to version 1.1 in 2019 and revised size thresholds for nonsolid nodules, added classification criteria for perifissural nodules, and allowed for short-interval follow-up of rapidly enlarging nodules that may be infectious in etiology. Lung-RADS v2022, released in November 2022, provides several updates including guidance on the classification and management of atypical pulmonary cysts, juxtapleural nodules, airway-centered nodules, and potentially infectious findings. This new release also provides clarification for determining nodule growth and introduces stepped management for nodules that are stable or decreasing in size. This article summarizes the current evidence and expert consensus supporting Lung-RADS v2022.

A novel image deep learning-based sub-centimeter pulmonary nodule management algorithm to expedite resection of the malignant and avoid over-diagnosis of the benign

OBJECTIVES

With the popularization of chest computed tomography (CT) screening, there are more sub-centimeter (≤ 1 cm) pulmonary nodules (SCPNs) requiring further diagnostic workup. This area represents an important opportunity to optimize the SCPN management algorithm avoiding "one-size fits all" approach. One critical problem is how to learn the discriminative multi-view characteristics and the unique context of each SCPN.

METHODS

Here, we propose a multi-view coupled self-attention module (MVCS) to capture the global spatial context of the CT image through modeling the association order of space and dimension. Compared with existing self-attention methods, MVCS uses less memory consumption and computational complexity, unearths dimension correlations that previous methods have not found, and is easy to integrate with other frameworks.

RESULTS

In total, a public dataset LUNA16 from LIDC-IDRI, 1319 SCPNs from 1069 patients presenting to a major referral center, and 160 SCPNs from 137 patients from three other major centers were analyzed to pre-train, train, and validate the model. Experimental results showed that performance outperforms the state-of-the-art models in terms of accuracy and stability and is comparable to that of human experts in classifying precancerous lesions and invasive adenocarcinoma. We also provide a fusion MVCS network (MVCSN) by combining the CT image with the clinical characteristics and radiographic features of patients.

CONCLUSION

This tool may ultimately aid in expediting resection of the malignant SCPNs and avoid over-diagnosis of the benign ones, resulting in improved management outcomes.

CLINICAL RELEVANCE STATEMENT

In the diagnosis of sub-centimeter lung adenocarcinoma, fusion MVCSN can help doctors improve work efficiency and guide their treatment decisions to a certain extent.

KEY POINTS

• Advances in computed tomography (CT) not only increase the number of nodules detected, but also the nodules that are identified are smaller, such as sub-centimeter pulmonary nodules (SCPNs). • We propose a multi-view coupled self-attention module (MVCS), which could model spatial and dimensional correlations sequentially for learning global spatial contexts, which is better than other attention mechanisms. • MVCS uses fewer huge memory consumption and computational complexity than the existing self-attention methods when dealing with 3D medical image data. Additionally, it reaches promising accuracy for SCPNs' malignancy evaluation and has lower training cost than other models.

Cooking oil fume exposure and Lung-RADS distribution among school cafeteria workers of South Korea

Background

Cooking oil fumes (COFs) from cooking with hot oil may contribute to the pathogenesis of lung cancer. Since 2021, occupational lung cancer for individual cafeteria workers has been recognized in South Korea. In this study, we aimed to identify the distribution of lung-imaging reporting and data system (Lung-RADS) among cafeteria workers and to determine factors related to Lung-RADS distribution.

Methods

We included 203 female participants who underwent low-dose computed tomography (LDCT) screening at a university hospital and examined the following variables: age, smoking status, second-hand smoke, height, weight, and years of service, mask use, cooking time, heat source, and ventilation. We divided all participants into culinary and non-culinary workers. Binomial logistic regression was conducted to determine the risk factors on LDCT of Category ≥ 3, separately for the overall group and the culinary group.

Results

In this study, Lung-RADS-positive occurred in 17 (8.4%) individuals, all of whom were culinary workers. Binary logistic regression analyses were performed and no variables were found to have a significant impact on Lung-RADS results. In the subgroup analysis, the Lung-RADS-positive, and -negative groups differed only in ventilation. Binary logistic regression showed that the adjusted odds ratio (aOR) of the Lung-RADS-positive group for inappropriate ventilation at the workplace was 14.89 (95% confidence interval [CI]: 3.296-67.231) compared to appropriate ventilation as the reference, and the aOR for electric appliances at home was 4.59 (95% CI: 1.061-19.890) using liquid fuel as the reference.

Conclusions

The rate of Lung-RADS-positive was significantly higher among culinary workers who performed actual cooking tasks than among nonculinary workers. In addition, appropriate ventilation at the workplace made the LDCT results differ. More research is needed to identify factors that might influence LDCT findings among culinary workers, including those in other occupations.

Rates of Downstream Procedures and Complications Associated With Lung Cancer Screening in Routine Clinical Practice : A Retrospective Cohort Study

BACKGROUND

Lung cancer screening (LCS) using low-dose computed tomography (LDCT) reduces lung cancer mortality but can lead to downstream procedures, complications, and other potential harms. Estimates of these events outside NLST (National Lung Screening Trial) have been variable and lacked evaluation by screening result, which allows more direct comparison with trials.

OBJECTIVE

To identify rates of downstream procedures and complications associated with LCS.

DESIGN

Retrospective cohort study.

SETTING

5 U.S. health care systems.

PATIENTS

Individuals who completed a baseline LDCT scan for LCS between 2014 and 2018.

MEASUREMENTS

Outcomes included downstream imaging, invasive diagnostic procedures, and procedural complications. For each, absolute rates were calculated overall and stratified by screening result and by lung cancer detection, and positive and negative predictive values were calculated.

RESULTS

Among the 9266 screened patients, 1472 (15.9%) had a baseline LDCT scan showing abnormalities, of whom 140 (9.5%) were diagnosed with lung cancer within 12 months (positive predictive value, 9.5% [95% CI, 8.0% to 11.0%]; negative predictive value, 99.8% [CI, 99.7% to 99.9%]; sensitivity, 92.7% [CI, 88.6% to 96.9%]; specificity, 84.4% [CI, 83.7% to 85.2%]). Absolute rates of downstream imaging and invasive procedures in screened patients were 31.9% and 2.8%, respectively. In patients undergoing invasive procedures after abnormal findings, complication rates were substantially higher than those in NLST (30.6% vs. 17.7% for any complication; 20.6% vs. 9.4% for major complications).

LIMITATION

Assessment of outcomes was retrospective and was based on procedural coding.

CONCLUSION

The results indicate substantially higher rates of downstream procedures and complications associated with LCS in practice than observed in NLST. Diagnostic management likely needs to be assessed and improved to ensure that screening benefits outweigh potential harms.

PRIMARY FUNDING SOURCE

National Cancer Institute and Gordon and Betty Moore Foundation.

Impact of Voxel Normalization on a Machine Learning-Based Method: A Study on Pulmonary Nodule Malignancy Diagnosis Using Low-Dose Computed Tomography (LDCT)

Lung cancer (LC) stands as the foremost cause of cancer-related fatality rates worldwide. Early diagnosis significantly enhances patient survival rate. Nowadays, low-dose computed tomography (LDCT) is widely employed on the chest as a tool for large-scale lung cancer screening. Nonetheless, a large amount of chest radiographs creates an onerous burden for radiologists. Some computer-aided diagnostic (CAD) tools can provide insight to the use of medical images for diagnosis and can augment diagnostic speed. However, due to the variation in the parameter settings across different patients, substantial discrepancies in image voxels persist. We found that different voxel sizes can create a compromise between model generalization and diagnostic efficacy. This study investigates the performance disparities of diagnostic models trained on original images and LDCT images reconstructed to different voxel sizes while making isotropic. We examined the ability of our method to differentiate between benign and malignant nodules. Using 11 features, a support vector machine (SVM) was trained on LDCT images using an isotropic voxel with a side length of 1.5 mm for 225 patients in-house. The result yields a favorable model performance with an accuracy of 0.9596 and an area under the receiver operating characteristic curve (ROC/AUC) of 0.9855. In addition, to furnish CAD tools for clinical application, future research including LDCT images from multi-centers is encouraged.

Lung cancer screening adherence among people living with and without HIV: An analysis of an integrated health system in Florida, United States (2012-2021)

Although lung cancer is a leading cause of death among people living with HIV (PLWH), limited research exists characterizing real-world lung cancer screening adherence among PLWH. Our objective was to compare low-dose computed tomography (LDCT) adherence among PLWH to those without HIV treated at one integrated health system. Using the University of Florida's Health Integrated Data Repository (01/01/2012-10/31/2021), we identified PLWH with at least one LDCT procedure, using Current Procedural Terminology codes(S8032/G0297/71271). Lung cancer screening adherence was defined as a second LDCT based on the Lung Imaging Reporting and Data System (Lung-RADS®). Lung-RADS categories were extracted from radiology reports using a natural language processing system. PLWH were matched with 4 randomly selected HIV-negative patients based on (+/- 1 year) age, Lung-RADS category, and calendar year. Seventy-three PLWH and 292 matched HIV-negative adults with at least one LDCT were identified. PLWH were more likely to be male (66% vs.52%,p < 0.04), non-Hispanic Black (53% vs.23%,p < 0.001), and live in an area of high poverty (45% vs.31%,p < 0.001). PLWH were more likely to be diagnosed with lung cancer after first LDCT (8% vs.0%,p < 0.001). Seventeen percent of HIV-negative and 12% of PLWH were adherent to LDCT screenings. Only 25% of PLWH diagnosed with category 4A were adherent compared to 44% of HIV-negative. On multivariable analyses, those with older age (66-80 vs.50-64 years) and with either Medicaid, charity-based, or other government insurance (vs. Medicare) were less likely to be adherent to LDCT screenings. PLWH may have poorer adherence to LDCT compared to their HIV-negative counterparts.

Developing electronic clinical quality measures to assess the cancer diagnostic process

OBJECTIVE

Measures of diagnostic performance in cancer are underdeveloped. Electronic clinical quality measures (eCQMs) to assess quality of cancer diagnosis could help quantify and improve diagnostic performance.

MATERIALS AND METHODS

We developed 2 eCQMs to assess diagnostic evaluation of red-flag clinical findings for colorectal (CRC; based on abnormal stool-based cancer screening tests or labs suggestive of iron deficiency anemia) and lung (abnormal chest imaging) cancer. The 2 eCQMs quantified rates of red-flag follow-up in CRC and lung cancer using electronic health record data repositories at 2 large healthcare systems. Each measure used clinical data to identify abnormal results, evidence of appropriate follow-up, and exclusions that signified follow-up was unnecessary. Clinicians reviewed 100 positive and 20 negative randomly selected records for each eCQM at each site to validate accuracy and categorized missed opportunities related to system, provider, or patient factors.

RESULTS

We implemented the CRC eCQM at both sites, while the lung cancer eCQM was only implemented at the VA due to lack of structured data indicating level of cancer suspicion on most chest imaging results at Geisinger. For the CRC eCQM, the rate of appropriate follow-up was 36.0% (26 746/74 314 patients) in the VA after removing clinical exclusions and 41.1% at Geisinger (1009/2461 patients; P < .001). Similarly, the rate of appropriate evaluation for lung cancer in the VA was 61.5% (25 166/40 924 patients). Reviewers most frequently attributed missed opportunities at both sites to provider factors (84 of 157).

CONCLUSIONS

We implemented 2 eCQMs to evaluate the diagnostic process in cancer at 2 large health systems. Health care organizations can use these eCQMs to monitor diagnostic performance related to cancer.

Invasive Procedures Associated With Lung Cancer Screening in Clinical Practice

BACKGROUND

The harm associated with imaging abnormalities related to lung cancer screening (LCS) is not well documented, especially outside the clinical trial and academic setting.

RESEARCH QUESTION

What is the frequency of invasive procedures and complications associated with a community based LCS program, including procedures for false-positive and benign, but clinically important, incidental findings?

STUDY DESIGN AND METHODS

We performed a single-center retrospective study of an LCS program at a nonuniversity teaching hospital from 2016 through 2019 to identify invasive procedures prompted by LCS results, including their indication and complications.

RESULTS

Among 2,003 LCS participants, 58 patients (2.9%) received a diagnosis of lung cancer and 71 patients (3.5%) received a diagnosis of any malignancy. Invasive procedures were performed 160 times in 103 participants (5.1%), including 1.7% of those without malignancy. Eight invasive procedures (0.4% of participants), including four surgeries (12% of diagnostic lung resections), were performed for false-positive lung nodules. Only 1% of Lung Imaging Reporting and Data System category 4A nodules that proved benign were subject to an invasive procedure. Among those without malignancy, an invasive procedure was performed in eight participants for extrapulmonary false-positive findings (0.4%) and in 19 participants (0.9%) to evaluate incidental findings considered benign but clinically important. Procedures for the latter indication resulted in treatment, change in management, or diagnosis in 79% of individuals. Invasive procedures in those without malignancy resulted in three complications (0.15%). Seventy nonsurgical procedures (6% complication rate) and 48 thoracic surgeries (4% major complication rate) were performed in those with malignancy.

INTERPRETATION

The use of invasive procedures to resolve false-positive findings was uncommon in the clinical practice of a nonuniversity LCS program that adhered to a nodule management algorithm and used a multidisciplinary approach. Incidental findings considered benign but clinically important resulted in invasive procedure rates that were similar to those for false-positive findings and frequently had clinical value.

"Will I change nodule management recommendations if I change my CAD system?"-impact of volumetric deviation between different CAD systems on lesion management

OBJECTIVES

To evaluate and compare the measurement accuracy of two different computer-aided diagnosis (CAD) systems regarding artificial pulmonary nodules and assess the clinical impact of volumetric inaccuracies in a phantom study.

METHODS

In this phantom study, 59 different phantom arrangements with 326 artificial nodules (178 solid, 148 ground-glass) were scanned at 80 kV, 100 kV, and 120 kV. Four different nodule diameters were used: 5 mm, 8 mm, 10 mm, and 12 mm. Scans were analyzed by a deep-learning (DL)-based CAD and a standard CAD system. Relative volumetric errors (RVE) of each system vs. ground truth and the relative volume difference (RVD) DL-based vs. standard CAD were calculated. The Bland-Altman method was used to define the limits of agreement (LOA). The hypothetical impact on LungRADS classification was assessed for both systems.

RESULTS

There was no difference between the three voltage groups regarding nodule volumetry. Regarding the solid nodules, the RVE of the 5-mm-, 8-mm-, 10-mm-, and 12-mm-size groups for the DL CAD/standard CAD were 12.2/2.8%, 1.3/ - 2.8%, - 3.6/1.5%, and - 12.2/ - 0.3%, respectively. The corresponding values for the ground-glass nodules (GGN) were 25.6%/81.0%, 9.0%/28.0%, 7.6/20.6%, and 6.8/21.2%. The mean RVD for solid nodules/GGN was 1.3/ - 15.2%. Regarding the LungRADS classification, 88.5% and 79.8% of all solid nodules were correctly assigned by the DL CAD and the standard CAD, respectively. 14.9% of the nodules were assigned differently between the systems.

CONCLUSIONS

Patient management may be affected by the volumetric inaccuracy of the CAD systems and hence demands supervision and/or manual correction by a radiologist.

KEY POINTS

• The DL-based CAD system was more accurate in the volumetry of GGN and less accurate regarding solid nodules than the standard CAD system. • Nodule size and attenuation have an effect on the measurement accuracy of both systems; tube voltage has no effect on measurement accuracy. • Measurement inaccuracies of CAD systems can have an impact on patient management, which demands supervision by radiologists.

ACR Lung CT Screening Reporting and Data System, a Systematic Review and Meta-Analysis Before Change in US Preventative Services Taskforce Eligibility Criteria: 2014 to 2021

OBJECTIVE

To review Lung CT Screening Reporting and Data System (Lung-RADS) scores from 2014 to 2021, before changes in eligibility criteria proposed by the US Preventative Services Taskforce.

METHODS

A registered systematic review and meta-analysis was conducted in MEDLINE, Embase, CINAHL, and Web of Science in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines; eligible studies examined low-dose CT (LDCT) lung cancer screening at institutions in the United States and reported Lung-RADS from 2014 to 2021. Patient and study characteristics, including age, gender, smoking status, pack-years, screening timeline, number of individual patients, number of unique studies, Lung-RADS scores, and positive predictive value (PPV) were extracted. Meta-analysis estimates were derived from generalized linear mixed modeling.

RESULTS

The meta-analysis included 24 studies yielding 36,211 LDCT examinations for 32,817 patient encounters. The meta-analysis Lung-RADS 1-2 scores were lower than anticipated by ACR guidelines, at 84.4 (95% confidence interval [CI] 83.3-85.6) versus 90% respectively (P < .001). Lung-RADS 3 and 4 scores were both higher than anticipated by the ACR, at 8.7% (95% CI 7.6-10.1) and 6.5% (95% CI 5.707.4), compared with 5% and 4%, respectively (P < .001). The ACR's minimum estimate of PPV for Lung-RADS 3 to 4 is 21% or higher; we observed a rate of 13.1% (95% CI 10.1-16.8). However, our estimated PPV rate for Lung-RADS 4 was 28.6% (95% CI 21.6-36.8).

CONCLUSION

Lung-RADS scores and PPV rates in the literature are not aligned with the ACR's own estimates, suggesting that perhaps Lung-RADS categorization needs to be reexamined for better concordance with real-world screening populations. In addition to serving as a benchmark before screening guideline broadening, this study provides guidance for future reporting of lung cancer screening and Lung-RADS data.

Multi-Omic Biomarkers Improve Indeterminate Pulmonary Nodule Malignancy Risk Assessment

The clinical management of patients with indeterminate pulmonary nodules is associated with unintended harm to patients and better methods are required to more precisely quantify lung cancer risk in this group. Here, we combine multiple noninvasive approaches to more accurately identify lung cancer in indeterminate pulmonary nodules. We analyzed 94 quantitative radiomic imaging features and 41 qualitative semantic imaging variables with molecular biomarkers from blood derived from an antibody-based microarray platform that determines protein, cancer-specific glycan, and autoantibody-antigen complex content with high sensitivity. From these datasets, we created a PSR (plasma, semantic, radiomic) risk prediction model comprising nine blood-based and imaging biomarkers with an area under the receiver operating curve (AUROC) of 0.964 that when tested in a second, independent cohort yielded an AUROC of 0.846. Incorporating known clinical risk factors (age, gender, and smoking pack years) for lung cancer into the PSR model improved the AUROC to 0.897 in the second cohort and was more accurate than a well-characterized clinical risk prediction model (AUROC = 0.802). Our findings support the use of a multi-omics approach to guide the clinical management of indeterminate pulmonary nodules.

Sybil: A Validated Deep Learning Model to Predict Future Lung Cancer Risk From a Single Low-Dose Chest Computed Tomography

PURPOSE

Low-dose computed tomography (LDCT) for lung cancer screening is effective, although most eligible people are not being screened. Tools that provide personalized future cancer risk assessment could focus approaches toward those most likely to benefit. We hypothesized that a deep learning model assessing the entire volumetric LDCT data could be built to predict individual risk without requiring additional demographic or clinical data.

METHODS

We developed a model called Sybil using LDCTs from the National Lung Screening Trial (NLST). Sybil requires only one LDCT and does not require clinical data or radiologist annotations; it can run in real time in the background on a radiology reading station. Sybil was validated on three independent data sets: a heldout set of 6,282 LDCTs from NLST participants, 8,821 LDCTs from Massachusetts General Hospital (MGH), and 12,280 LDCTs from Chang Gung Memorial Hospital (CGMH, which included people with a range of smoking history including nonsmokers).

RESULTS

Sybil achieved area under the receiver-operator curves for lung cancer prediction at 1 year of 0.92 (95% CI, 0.88 to 0.95) on NLST, 0.86 (95% CI, 0.82 to 0.90) on MGH, and 0.94 (95% CI, 0.91 to 1.00) on CGMH external validation sets. Concordance indices over 6 years were 0.75 (95% CI, 0.72 to 0.78), 0.81 (95% CI, 0.77 to 0.85), and 0.80 (95% CI, 0.75 to 0.86) for NLST, MGH, and CGMH, respectively.

CONCLUSION

Sybil can accurately predict an individual's future lung cancer risk from a single LDCT scan to further enable personalized screening. Future study is required to understand Sybil's clinical applications. Our model and annotations are publicly available.

[Media: see text].

Lung Cancer Screening Using Clinical Photon-Counting Detector Computed Tomography and Energy-Integrating-Detector Computed Tomography: A Prospective Patient Study

OBJECTIVE

To evaluate the diagnostic quality of photon-counting detector (PCD) computed tomography (CT) in patients undergoing lung cancer screening compared with conventional energy-integrating detector (EID) CT in a prospective multireader study.

MATERIALS

Patients undergoing lung cancer screening with conventional EID-CT were prospectively enrolled and scanned on a PCD-CT system using similar automatic exposure control settings and reconstruction kernels. Three thoracic radiologists blinded to CT system compared PCD-CT and EID-CT images and scored examinations using a 5-point Likert comparison score (-2 [left image is worse] to +2 [left image is better]) for artifacts, sharpness, image noise, diagnostic image quality, emphysema visualization, and lung nodule evaluation focusing on the border. Post hoc correction of Likert scores was performed such that they reflected PCD-CT performance in comparison to EID-CT. A nonreader radiologist measured objective image noise.

RESULTS

Thirty-three patients (mean, 66.9 ± 5.6 years; 11 female; body mass index; 30.1 ± 5.1 kg/m 2 ) were enrolled. Mean volume CT dose index for PCD-CT was lower (0.61 ± 0.21 vs 0.73 ± 0.22; P < 0.001). Pooled reader results showed significant differences between imaging modalities for all comparative rankings ( P < 0.001), with PCD-CT favored for sharpness, image noise, image quality, and emphysema visualization and lung nodule border, but not artifacts. Photon-counting detector CT had significantly lower image noise (74.4 ± 10.5 HU vs 80.1 ± 8.6 HU; P = 0.048).

CONCLUSIONS

Photon-counting detector CT with similar acquisition and reconstruction settings demonstrated improved image quality and less noise despite lower radiation dose, with improved ability to depict pulmonary emphysema and lung nodule borders compared with EID-CT at low-dose lung cancer CT screening.

Clinical Scores, Biomarkers and IT Tools in Lung Cancer Screening-Can an Integrated Approach Overcome Current Challenges?

As most lung cancer (LC) cases are still detected at advanced and incurable stages, there are increasing efforts to foster detection at earlier stages by low dose computed tomography (LDCT) based LC screening. In this scoping review, we describe current advances in candidate selection for screening (selection phase), technical aspects (screening), and probability evaluation of malignancy of CT-detected pulmonary nodules (PN management). Literature was non-systematically assessed and reviewed for suitability by the authors. For the selection phase, we describe current eligibility criteria for screening, along with their limitations and potential refinements through advanced clinical scores and biomarker assessments. For LC screening, we discuss how the accuracy of computerized tomography (CT) scan reading might be augmented by IT tools, helping radiologists to cope with increasing workloads. For PN management, we evaluate the precision of follow-up scans by semi-automatic volume measurements of CT-detected PN. Moreover, we present an integrative approach to evaluate the probability of PN malignancy to enable safe decisions on further management. As a clear limitation, additional validation studies are required for most innovative diagnostic approaches presented in this article, but the integration of clinical risk models, current imaging techniques, and advancing biomarker research has the potential to improve the LC screening performance generally.

[Lung Imaging Reporting and Data System (Lung-RADS) in Radiology: Strengths, Weaknesses and Improvement]

In 2019, the American College of Radiology announced Lung CT Screening Reporting & Data System (Lung-RADS) 1.1 to reduce lung cancer false positivity compared to that of Lung-RADS 1.0 for effective national lung cancer screening, and in December 2022, announced the new Lung-RADS 1.1, Lung-RADS^® 2022 improvement. The Lung-RADS^® 2022 measures the nodule size to the first decimal place compared to that of the Lung-RADS 1.0, to category 2 until the juxtapleural nodule size is < 10 mm, increases the size criterion of the ground glass nodule to 30 mm in category 2, and changes categories 4B and 4X to extremely suspicious. The category was divided according to the airway nodules location and shape or wall thickness of atypical pulmonary cysts. Herein, to help radiologists understand the Lung-RADS^® 2022, this review will describe its advantages, disadvantages, and future improvements.

Quantitative parameters of enhanced dual-energy computed tomography for differentiating lung cancers from benign lesions in solid pulmonary nodules

Objectives

This study aimed to investigate the ability of quantitative parameters of dual-energy computed tomography (DECT) and nodule size for differentiation between lung cancers and benign lesions in solid pulmonary nodules.

Materials and Methods

A total of 151 pathologically confirmed solid pulmonary nodules including 78 lung cancers and 73 benign lesions from 147 patients were consecutively and retrospectively enrolled who underwent dual-phase contrast-enhanced DECT. The following features were analyzed: diameter, volume, Lung CT Screening Reporting and Data System (Lung-RADS) categorization, and DECT-derived quantitative parameters including effective atomic number (Zeff), iodine concentration (IC), and normalized iodine concentration (NIC) in arterial and venous phases. Multivariable logistic regression analysis was used to build a combined model. The diagnostic performance was assessed by area under curve (AUC) of receiver operating characteristic curve, sensitivity, and specificity.

Results

The independent factors for differentiating lung cancers from benign solid pulmonary nodules included diameter, Lung-RADS categorization of diameter, volume, Zeff in arterial phase (Zeff_A), IC in arterial phase (IC_A), NIC in arterial phase (NIC_A), Zeff in venous phase (Zeff_V), IC in venous phase (IC_V), and NIC in venous phase (NIC_V) (all P &lt; 0.05). The IC_V, NIC_V, and combined model consisting of diameter and NIC_V showed good diagnostic performance with AUCs of 0.891, 0.888, and 0.893, which were superior to the diameter, Lung-RADS categorization of diameter, volume, Zeff_A, and Zeff_V (all P &lt; 0.001). The sensitivities of IC_V, NIC_V, and combined model were higher than those of IC_A and NIC_A (all P &lt; 0.001). The combined model did not increase the AUCs compared with IC_V (P = 0.869) or NIC_V (P = 0.633).

Conclusion

The DECT-derived IC_V and NIC_V may be useful in differentiating lung cancers from benign lesions in solid pulmonary nodules.

Intra- and Inter-Reader Variations in Lung Nodule Measurements: Influences of Nodule Size, Location, and Observers

(1) Background: Accurate measurement of lung-nodule size is necessary, but whether a three-dimensional volume measurement is better or more reliable than the one-dimensional method is still unclear. This study aimed to investigate the intra- and inter-reader variations according to nodule type, size, three-dimensional volume measurements, and one-dimensional linear measurements. (2) Methods: This retrospective study included computed tomography (CT) examinations of lung nodules and volume measurements performed from October to December 2016. Two radiologists independently performed all measurements. Intra-class correlation coefficients (ICC) and Bland-Altman plots were used for analysis. (3) Results: The overall variability in the calculated volume was larger than when using the semiautomatic volume measurement. Nodules <6 mm tended to have larger variability than nodules ≥6 mm in both one-dimensional and calculated volume measurements. The isolated type showed smaller variability in both intra- and inter-reader comparisons. The juxta-vascular type showed the largest variability in both one-dimensional and calculated volume measurements. The variability was decreased when using the 3D volume semiautomated software. (4) Conclusions: The present study suggests that 3D semiautomatic volume measurements showed lower variability than the calculated volume measurement. Nodule size and location influence measurement variability. The intra- and inter-reader variabilities in nodule volume measurement were considerable.

Strategies for Reducing False-Positive Screening Results for Intermediate-Size Nodules Evaluated Using Lung-RADS: A Secondary Analysis of National Lung Screening Trial Data

BACKGROUND. Lung-RADS version 1.1 (v1.1) classifies all solid nodules less than 6 mm as category 2. Lung-RADS v1.1 also classifies solid intermediate-size (6 to < 10 mm) nodules as category 2 if they are perifissural and have a triangular, polygonal, or ovoid shape (indicative of intrapulmonary lymph nodes). Additional category 2 criteria could reduce false-positive results of screening examinations. OBJECTIVE. The purpose of this study was to evaluate the impact of proposed strategies for reducing false-positive results for intermediate-size nodules on lung cancer screening CT evaluated using Lung-RADS v1.1. METHODS. This retrospective study entailed secondary analysis of National Lung Screening Trial (NLST) data. Of 1387 solid nodules measuring 6.0-9.5 mm on baseline screening CT examinations in the NLST, all 38 nodules in patients who developed cancer and a random sample of 200 nodules in patients who did not develop cancer were selected for further evaluation. Cancers were required to correspond with the baseline nodule on manual review. After exclusions, the sample included 223 patients (median age, 62 years; 143 men, 80 women; 196 benign nodules, 27 malignant nodules). Two thoracic radiologists independently reviewed baseline examinations to record nodule diameter and volume using semiautomated software and to determine whether nodules had perifissural location; other subpleural location; and triangular, polygonal, or ovoid shape. Different schemes for category 2 assignment were compared. RESULTS. Across readers, standard Lung-RADS v1.1 had sensitivity of 89-93% and specificity of 26-31%. A modification assigning nodules less than 10 mm with triangular, polygonal, or ovoid shape in other subpleural locations (vs only perifissural location) as category 2 had sensitivity of 85-93% and specificity of 47-51%. Lung-RADS v1.1 using volume cutoffs had sensitivity of 89-93% and specificity of 37% (both readers). The sensitivity of both modified Lung-RADS v1.1 and Lung-RADS v1.1 with volume cutoffs was not significantly different from standard Lung-RADS v1.1 (all p > .05). However, both schemes' specificity was significantly better than standard Lung-RADS v1.1 (all p < .05). Combining the two strategies yielded sensitivity of 85-93% and specificity of 58-59%. CONCLUSION. Classifying intermediate-size nodules with triangular, polygonal, or ovoid shape in any subpleural (not just perifissural) location as category 2 and using volume- rather than diameter-based measurements improves Lung-RADS specificity without decreased sensitivity. CLINICAL IMPACT. The findings can help reduce false-positive results, decreasing 6-month follow-up examinations for benign findings.

CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Radiomics and malignancy prediction

Spiculations/lobulations, sharp/curved spikes on the surface of lung nodules, are good predictors of lung cancer malignancy and hence, are routinely assessed and reported by radiologists as part of the standardized Lung-RADS clinical scoring criteria. Given the 3D geometry of the nodule and 2D slice-by-slice assessment by radiologists, manual spiculation/lobulation annotation is a tedious task and thus no public datasets exist to date for probing the importance of these clinically-reported features in the SOTA malignancy prediction algorithms. As part of this paper, we release a large-scale Clinically-Interpretable Radiomics Dataset, CIRDataset, containing 956 radiologist QA/QC'ed spiculation/lobulation annotations on segmented lung nodules from two public datasets, LIDC-IDRI (N=883) and LUNGx (N=73). We also present an end-to-end deep learning model based on multi-class Voxel2Mesh extension to segment nodules (while preserving spikes), classify spikes (sharp/spiculation and curved/lobulation), and perform malignancy prediction. Previous methods have performed malignancy prediction for LIDC and LUNGx datasets but without robust attribution to any clinically reported/actionable features (due to known hyperparameter sensitivity issues with general attribution schemes). With the release of this comprehensively-annotated CIRDataset and end-to-end deep learning baseline, we hope that malignancy prediction methods can validate their explanations, benchmark against our baseline, and provide clinically-actionable insights. Dataset, code, pretrained models, and docker containers are available at https://github.com/nadeemlab/CIR.

Evaluation of benefits and harms of adaptive screening schedules for lung cancer: A microsimulation study

BACKGROUND

Although lung cancer screening (LCS) has been proven effective in reducing lung cancer mortality, it is associated with some potential harms, such as false positives and invasive follow-up procedures. Determining the time to next screen based on individual risk could reduce harms while maintaining health gains. Here, we evaluate the benefits and harms of LCS strategies with adaptive schedules, and compare these with those from non-adaptive strategies.

METHODS

We extended the Lee and Zelen risk threshold method to select screening schedules based on individual's lung cancer risk and life expectancy (adaptive schedules). We compared the health benefits and harms of these adaptive schedules with regular (non-adaptive) schedules (annual, biennial and triennial) using a validated lung cancer microsimulation model. Outcomes include lung cancer deaths (LCD) averted, life years gained (LYG), discounted quality adjusted life years (QALYs) gained, and false positives per LCD averted. We also explored the impact of varying screening-related disutilities.

RESULTS

In comparison to standard regular screening recommendations, risk-dependent adaptive screening reduced screening harms while maintaining a similar level of health benefits. The net gains and the balance of benefits and harms from LCS with efficient adaptive schedules were improved compared to those from regular screening, especially when the screening-related disutilities are high.

CONCLUSIONS

Adaptive screening schedules can reduce the associated harms of screening while maintaining its associated lung cancer mortality reductions and years of life gained. Our study identifies individually tailored schedules that optimize the screening benefit/harm trade-offs.

NCCN Guidelines® Insights: Lung Cancer Screening, Version 1.2022

The NCCN Guidelines for Lung Cancer Screening recommend criteria for selecting individuals for screening and provide recommendations for evaluation and follow-up of lung nodules found during initial and subsequent screening. These NCCN Guidelines Insights focus on recent updates to the NCCN Guidelines for Lung Cancer Screening.

Low-dose CT for lung cancer screening: position paper from the Italian college of thoracic radiology

Smoking is the main risk factor for lung cancer (LC), which is the leading cause of cancer-related death worldwide. Independent randomized controlled trials, governmental and inter-governmental task forces, and meta-analyses established that LC screening (LCS) with chest low dose computed tomography (LDCT) decreases the mortality of LC in smokers and former smokers, compared to no-screening, especially in women. Accordingly, several Italian initiatives are offering LCS by LDCT and smoking cessation to about 10,000 high-risk subjects, supported by Private or Public Health Institutions, envisaging a possible population-based screening program. Because LDCT is the backbone of LCS, Italian radiologists with LCS expertise are presenting this position paper that encompasses recommendations for LDCT scan protocol and its reading. Moreover, fundamentals for classification of lung nodules and other findings at LDCT test are detailed along with international guidelines, from the European Society of Thoracic Imaging, the British Thoracic Society, and the American College of Radiology, for their reporting and management in LCS. The Italian College of Thoracic Radiologists produced this document to provide the basics for radiologists who plan to set up or to be involved in LCS, thus fostering homogenous evidence-based approach to the LDCT test over the Italian territory and warrant comparison and analyses throughout National and International practices.

The impact of cardiopulmonary hemodynamic factors in volumetry for pulmonary nodule management

BACKGROUND

The acceptance of coronary CT angiogram (CCTA) scans in the management of stable angina has led to an exponential increase in studies performed and reported incidental findings, including pulmonary nodules (PN). Using low-dose CT scans, volumetry tools are used in growth assessment and risk stratification of PN between 5 and 8 mm in diameter. Volumetry of PN could also benefit from the increased temporal resolution of CCTA scans, potentially expediting clinical decisions when an incidental PN is first detected on a CCTA scan, and allow for better resource management and planning in a Radiology department. This study aims to investigate how cardiopulmonary hemodynamic factors impact the volumetry of PN using CCTA scans. These factors include the cardiac phase, vascular distance from the main pulmonary artery (MPA) to the nodule, difference of the MPA diameter between systole and diastole, nodule location, and cardiomegaly presence.

MATERIALS AND METHODS

Two readers reviewed all CCTA scans performed from 2016 to 2019 in a tertiary hospital and detected PN measuring between 5 and 8 mm in diameter. Each observer measured each nodule using two different software packages and in systole and diastole. A multiple linear regression model was applied, and inter-observer and inter-software agreement were assessed using intraclass correlation.

RESULTS

A total of 195 nodules from 107 patients were included in this retrospective, cross-sectional and observational study. The regression model identified the vascular distance (p < 0.001), the difference of the MPA diameter between systole and diastole (p < 0.001), and the location within the lower or posterior thirds of the field of view (p < 0.001 each) as affecting the volume measurement. The cardiac phase was not significant in the model. There was a very high inter-observer agreement but no reasonable inter-software agreement between measurements.

CONCLUSIONS

PN volumetry using CCTA scans seems to be sensitive to cardiopulmonary hemodynamic changes independently of the cardiac phase. These might also be relevant to non-gated scans, such as during PN follow-up. The cardiopulmonary hemodynamic changes are a new limiting factor to PN volumetry. In addition, when a patient experiences an acute or deteriorating cardiopulmonary disease during PN follow-up, these hemodynamic changes could affect the PN growth estimation.

Personalized Chest Computed Tomography: Minimum Diagnostic Radiation Dose Levels for the Detection of Fibrosis, Nodules, and Pneumonia

OBJECTIVES

The purpose of this study was to evaluate the minimum diagnostic radiation dose level for the detection of high-resolution (HR) lung structures, pulmonary nodules (PNs), and infectious diseases (IDs).

MATERIALS AND METHODS

A preclinical chest computed tomography (CT) trial was performed with a human cadaver without known lung disease with incremental radiation dose using tin filter-based spectral shaping protocols. A subset of protocols for full diagnostic evaluation of HR, PN, and ID structures was translated to clinical routine. Also, a minimum diagnostic radiation dose protocol was defined (MIN). These protocols were prospectively applied over 5 months in the clinical routine under consideration of the individual clinical indication. We compared radiation dose parameters, objective and subjective image quality (IQ).

RESULTS

The HR protocol was performed in 38 patients (43%), PN in 21 patients (24%), ID in 20 patients (23%), and MIN in 9 patients (10%). Radiation dose differed significantly among HR, PN, and ID (5.4, 1.2, and 0.6 mGy, respectively; P < 0.001). Differences between ID and MIN (0.2 mGy) were not significant (P = 0.262). Dose-normalized contrast-to-noise ratio was comparable among all groups (P = 0.087). Overall IQ was perfect for the HR protocol (median, 5.0) and decreased for PN (4.5), ID-CT (4.3), and MIN-CT (2.5). The delineation of disease-specific findings was high in all dedicated protocols (HR, 5.0; PN, 5.0; ID, 4.5). The MIN protocol had borderline IQ for PN and ID lesions but was insufficient for HR structures. The dose reductions were 78% (PN), 89% (ID), and 97% (MIN) compared with the HR protocols.

CONCLUSIONS

Personalized chest CT tailored to the clinical indications leads to substantial dose reduction without reducing interpretability. More than 50% of patients can benefit from such individual adaptation in a clinical routine setting. Personalized radiation dose adjustments with validated diagnostic IQ are especially preferable for evaluating ID and PN lesions.

Incidental findings on lung cancer screening: pictorial essay and systematic checklist

Lung cancer screening (LCS) programs are increasing worldwide. Incidental findings (IFs) on LCS are defined as low-dose CT findings unrelated to the primary purpose of identifying lung cancer. Most IFs on LCS are benign and clinically insignificant but are being increasingly recognized, and some require urgent referral for further diagnostic workup. Other findings are expected and are known as smoking-related comorbidities, including COPD, cardiovascular disease, emphysema, and interstitial lung disease, and their diagnosis can have a significant impact on patient prognosis. The purpose of this pictorial essay is to illustrate the most common IFs on LCS, organized by organ. We will discuss the current literature on IFs on LCS, focusing on their prevalence, appropriate communication, and triggering of clinical pathway systems.

Efficacy of Ultrashort Echo Time Pulmonary MRI for Lung Nodule Detection and Lung-RADS Classification

Background Pulmonary MRI with ultrashort echo time (UTE) has been compared with chest CT for nodule detection and classification. However, direct comparisons of these methods' capabilities for Lung CT Screening Reporting and Data System (Lung-RADS) evaluation remain lacking. Purpose To compare the capabilities of pulmonary MRI with UTE with those of standard- or low-dose thin-section CT for Lung-RADS classification. Materials and Methods In this prospective study, standard- and low-dose chest CT (270 mA and 60 mA, respectively) and MRI with UTE were used to examine consecutive participants enrolled between January 2017 and December 2020 who met American College of Radiology Appropriateness Criteria for lung cancer screening with low-dose CT. Probability of nodule presence was assessed for all methods with a five-point visual scoring system by two board-certified radiologists. All nodules were then evaluated in terms of their Lung-RADS classification using each method. To compare nodule detection capability of the three methods, consensus for performances was rated by using jackknife free-response receiver operating characteristic analysis, and sensitivity was compared by means of the McNemar test. In addition, weighted κ statistics were used to determine the agreement between Lung-RADS classification obtained with each method and the reference standard generated from standard-dose CT evaluated by two radiologists who were not included in the image analysis session. Results A total of 205 participants (mean age: 64 years ± 7 [standard deviation], 106 men) with 1073 nodules were enrolled. Figure of merit (FOM) (P < .001) had significant differences among three modalities (standard-dose CT: FOM = 0.91, low-dose CT: FOM = 0.89, pulmonary MRI with UTE: FOM = 0.94), with no evidence of false-positive findings in participants with all modalities (P > .05). Agreements for Lung-RADS classification between all modalities and the reference standard were almost perfect (standard-dose CT: κ = 0.82, P < .001; low-dose CT: κ = 0.82, P < .001; pulmonary MRI with UTE: κ = 0.82, P < .001). Conclusion In a lung cancer screening population, ultrashort echo time pulmonary MRI was comparable to standard- or low-dose CT for Lung CT Screening Reporting and Data System classification. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Wielpütz in this issue.

Outcomes of Positive and Suspicious Findings in Clinical CT Lung Cancer Screening and the Road Ahead

Rationale

Future optimization of computed tomography (CT) lung cancer screening (CTLS) algorithms will depend on clinical outcomes data.

Objectives

To report the outcomes of positive and suspicious findings in a clinical CTLS program.

Methods

We retrospectively reviewed results for patients from our institution undergoing lung cancer screening from January 2012 through December 2018, with follow-up through December 2019. All exams were retrospectively rescored using Lung-RADS v1.1 (LR). Metrics assessed included positive, probably benign, and suspicious exam rates, frequency/nature of care escalation, and lung cancer detection rates after a positive, probably benign, and suspicious exam result and overall. We calculated time required to resolve suspicious exams as malignant or benign. Results were broken down by subcategories, reason for positive/suspicious designation, and screening round.

Results

During the study period 4,301 individuals underwent a total of 10,897 exams. The number of positive (13.9%), suspicious (5.5%), and significant incidental (6.4%) findings was significantly higher at baseline screening. Cancer detection and false-positive rates were 2.0% and 12.3% at baseline versus 1.3% and 5.1% across subsequent screening rounds, respectively. Baseline solid nodule(s) 6 to <8 mm were the only probably benign findings resulting in lung cancer detection within 12 months. New solid nodules 6 to <8 mm were the only LR category 4A (LR4A) findings falling within the LR predicted cancer detection range of 5–15% (12.8%). 38.5% of LR4A cancers were detected within 3 months.

Conclusions

Modification of the definition and suggested workup of positive and suspicious lung cancer screening findings appears warranted.

AI-Based Detection, Classification and Prediction/Prognosis in Medical Imaging:: Towards Radiophenomics

Artificial intelligence (AI) techniques have significant potential to enable effective, robust, and automated image phenotyping including the identification of subtle patterns. AI-based detection searches the image space to find the regions of interest based on patterns and features. There is a spectrum of tumor histologies from benign to malignant that can be identified by AI-based classification approaches using image features. The extraction of minable information from images gives way to the field of "radiomics" and can be explored via explicit (handcrafted/engineered) and deep radiomics frameworks. Radiomics analysis has the potential to be used as a noninvasive technique for the accurate characterization of tumors to improve diagnosis and treatment monitoring. This work reviews AI-based techniques, with a special focus on oncological PET and PET/CT imaging, for different detection, classification, and prediction/prognosis tasks. We also discuss needed efforts to enable the translation of AI techniques to routine clinical workflows, and potential improvements and complementary techniques such as the use of natural language processing on electronic health records and neuro-symbolic AI techniques.