Lung Imaging Reporting And Data System Research Articles

Prognostic value of biohybrid screening results for assessing cancer risk using the example of lung cancer

Introduction. One of the leading places in the structure of socially significant diseases in the Russian Federation is occupied by malignant neoplasms, among which cancer of the trachea, bronchi and lung occupies the first place in terms of prevalence. Screening for lung cancer is not carried out at the federal level in Russia; recommended screening measures include questionnaires to identify risk factors for the development of the disease (including occupational) and the formation of dispensary observation groups. The study aims to assess the prognostic value of a new method of biohybrid screening of lung cancer by exhaled air. Materials and methods. Scientists conducted a study at the National Medical Research Center of Oncology of the Ministry of Health, Russian Federation, with the participation of 24 conditionally healthy volunteers and 5 patients diagnosed with lung cancer in the early stages of the disease. The design of the study is simultaneous (on-line) with blinding; experts estimated the risk of systematic errors at 10 points on the QUADAS scale. The authors examined exhaled air samples on a gas analyzer (biohybrid detector) of the BGS-APK 02SZZ model, in which the sensor were animals — gray domesticated rats (Rattus norvegicus) with microelectrode matrices implanted into the olfactory bulb, and which were in the internal (ventilated) space of the analyzer in a state of drug anesthesia. The conclusion about the presence or absence of a risk of lung cancer was formed by an artificial neural network of the analyzer, at the input of which bioelectric signals from each microelectrode were received, which were preprocessed using a system for recording the focal activity of the olfactory analyzer. The specialists confirmed the presence or absence of a risk of lung cancer by a reference method, according to computed tomography of the chest in accordance with the Lung Imaging Reporting and Data System (LungRADS™). Results. During the study, experts determined the effectiveness indicators of the method of biohybrid lung cancer screening by exhaled air: sensitivity (64%), specificity (87%), prognosticality of negative and positive conclusions about the presence/absence of disease risk (82% and 72%, respectively). Conclusion. The safety, efficiency and throughput of the investigated method, together with the reproducibility of the analyzer's conclusions, meet the criteria of practicability to ensure the first (population-based) stage of lung cancer screening.

Comparison of Lung-RADS Version 1.1 and Lung-RADS Version 2022 in Classifying Airway Nodules Detected at Lung Cancer Screening CT.

Purpose To compare the Lung Imaging Reporting and Data System (Lung-RADS) version 1.1 with version 2022 classification of airway nodules detected at lung cancer screening CT examinations. Materials and Methods This retrospective study included all patients who underwent a lung cancer screening CT examination in the authors' health care network between 2015 and 2021 with a reported airway or endobronchial nodule. A fellowship-trained cardiothoracic radiologist reviewed these CT images and characterized the airway nodules by size, location, multiplicity, morphology, dependent portions of airway, internal air, fluid attenuation, distal changes, outcome at follow-up, and final pathologic diagnosis, if malignant. Sensitivity and specificity of Lung-RADS version 1.1 in detecting malignant nodules were compared with those of Lung-RADS version 2022 using the McNemar test. Results A total of 174 patients were included. Of these, 163 (94%) had airway nodules that were deemed benign, while 11 (6%) had malignant nodules. Airway nodules in the trachea and mainstem bronchi were all benign, while lobar and segmental airway nodules had the highest risk for lung cancer (17.2% and 11.1%, respectively). Of the 12 subsegmental airway nodules that were obstructive, three (25%) were malignant and nine (75%) were benign. Nodules with nonobstructive morphologies, dependent portions of airway, internal air, or fluid attenuation were all benign. Only 10 of the 92 (10.9%) patients with positive Lung-RADS by clinical report had cancer. Lung-RADS version 2022 resulted in higher specificity than version 1.1 (82% vs 50%, P < .001), without sacrificing sensitivity (91% for both). Conclusion Compared with the previous version, Lung-RADS version 2022 reduced the number of false-positive screening CT examinations while still identifying malignant airway nodules. Keywords: CT, Lung, Primary Neoplasms, Pulmonary, Lung Cancer Screening, Lung-RADS, Nodule Risk, Airway Nodule, Endobronchial Nodule © RSNA, 2024.

Comparison of different classification systems for pulmonary nodules: a multicenter retrospective study in China

BackgroundTo compare the diagnostic performance of Lung-RADS (lung imaging-reporting and data system) 2022 and PNI-GARS (pulmonary node imaging-grading and reporting system).MethodsPulmonary nodules (PNs) were selected at four centers, namely, CQ Center (January 1, 2018-December 31, 2021), HB Center (January 1, 2021–June 30, 2022), SC Center (September 1, 2021–December 31, 2021), and SX Center (January 1, 2021–December 31, 2021). PNs were divided into solid nodules (SNs), partial solid nodules (PSNs) and ground-glass nodules (GGNs), and they were then classified by the Lung-RADS and PNI-GARS. The sensitivity, specificity and agreement rate were compared between the two systems by the χ2 test.ResultsFor SN and PSN, the sensitivity of PNI-GARS and Lung-RADS was close (SN 99.8% vs. 99.4%, P < 0.001; PSN 99.9% vs. 98.4%, P = 0.015), but the specificity (SN 51.2% > 35.1%, PSN 13.3% > 5.7%, all P < 0.001) and agreement rate (SN 81.1% > 74.5%, P < 0.001, PSN 94.6% > 92.7%, all P < 0.05) of PNI-GARS were superior to those of Lung-RADS. For GGN, the sensitivity (96.5%) and agreement rate (88.6%) of PNI-GARS were better than those of Lung-RADS (0, 18.5%, P < 0.001). For the whole sample, the sensitivity (98.5%) and agreement rate (87.0%) of PNI-GARS were better than Lung-RADS (57.5%, 56.5%, all P < 0.001), whereas the specificity was slightly lower (49.8% < 53.4%, P = 0.003).ConclusionPNI-GARS was superior to Lung-RADS in diagnostic performance, especially for GGN.

Combined model integrating deep learning, radiomics, and clinical data to classify lung nodules at chest CT.

The study aimed to develop a combined model that integrates deep learning (DL), radiomics, and clinical data to classify lung nodules into benign or malignant categories, and to further classify lung nodules into different pathological subtypes and Lung Imaging Reporting and Data System (Lung-RADS) scores. The proposed model was trained, validated, and tested using three datasets: one public dataset, the Lung Nodule Analysis 2016 (LUNA16) Grand challenge dataset (n = 1004), and two private datasets, the Lung Nodule Received Operation (LNOP) dataset (n = 1027) and the Lung Nodule in Health Examination (LNHE) dataset (n = 1525). The proposed model used a stacked ensemble model by employing a machine learning (ML) approach with an AutoGluon-Tabular classifier. The input variables were modified 3D convolutional neural network (CNN) features, radiomics features, and clinical features. Three classification tasks were performed: Task 1: Classification of lung nodules into benign or malignant in the LUNA16 dataset; Task 2: Classification of lung nodules into different pathological subtypes; and Task 3: Classification of Lung-RADS score. Classification performance was determined based on accuracy, recall, precision, and F1-score. Ten-fold cross-validation was applied to each task. The proposed model achieved high accuracy in classifying lung nodules into benign or malignant categories in LUNA 16 with an accuracy of 92.8%, as well as in classifying lung nodules into different pathological subtypes with an F1-score of 75.5% and Lung-RADS scores with an F1-score of 80.4%. Our proposed model provides an accurate classification of lung nodules based on the benign/malignant, different pathological subtypes, and Lung-RADS system.

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

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. 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. A total of 242 NSNs (mean diameter, 7.6 ± 2.8mm), 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. The window-based visual classification of NSNs is a simple and objective method to discriminate IADs from NILs. The present study shows that using the bone window to classify nonsolid nodules helps discriminate invasive adenocarcinoma from noninvasive lesions. • 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.

How AI Responds to Common Lung Cancer Questions: ChatGPT versus Google Bard

Background The recent release of large language models (LLMs) for public use, such as ChatGPT and Google Bard, has opened up a multitude of potential benefits as well as challenges. Purpose To evaluate and compare the accuracy and consistency of responses generated by publicly available ChatGPT-3.5 and Google Bard to non-expert questions related to lung cancer prevention, screening, and terminology commonly used in radiology reports based on the recommendation of Lung Imaging Reporting and Data System (Lung-RADS) v2022 from American College of Radiology and Fleischner society. Materials and Methods Forty of the exact same questions were created and presented to ChatGPT-3.5 and Google Bard experimental version as well as Bing and Google search engines by three different authors of this paper. Each answer was reviewed by two radiologists for accuracy. Responses were scored as correct, partially correct, incorrect, or unanswered. Consistency was also evaluated among the answers. Here, consistency was defined as the agreement between the three answers provided by ChatGPT-3.5, Google Bard experimental version, Bing, and Google search engines regardless of whether the concept conveyed was correct or incorrect. The accuracy among different tools were evaluated using Stata. Results ChatGPT-3.5 answered 120 questions with 85 (70.8%) correct, 14 (11.7%) partially correct, and 21 (17.5%) incorrect. Google Bard did not answer 23 (19.1%) questions. Among the 97 questions answered by Google Bard, 62 (51.7%) were correct, 11 (9.2%) were partially correct, and 24 (20%) were incorrect. Bing answered 120 questions with 74 (61.7%) correct, 13 (10.8%) partially correct, and 33 (27.5%) incorrect. Google search engine answered 120 questions with 66 (55%) correct, 27 (22.5%) partially correct, and 27 (22.5%) incorrect. The ChatGPT-3.5 is more likely to provide correct or partially answer than Google Bard, approximately by 1.5 folds (OR = 1.55, P = 0.004). ChatGPT-3.5 and Google search engine were more likely to be consistent than Google Bard by approximately 7 and 29 folds (OR = 6.65, P = 0.002 for ChatGPT and OR = 28.83, P = 0.002 for Google search engine, respectively). Conclusion Although ChatGPT-3.5 had a higher accuracy in comparison with the other tools, neither ChatGPT nor Google Bard, Bing and Google search engines answered all questions correctly and with 100% consistency.

The influence of different previous cancer histories on the diagnostic efficacy of Lung Imaging Reporting and Data System.

For lung cancer screening in patients with previous malignant tumors, Lung Imaging Reporting and Data System (Lung-RADS) and other lung cancer screening tools are controversial in terms of requirements for the previous cancer history. This study investigated the effect of the length and type of malignancy history on the diagnostic efficacy of Lung Imaging Reporting and Data System (Lung-RADS) 2022 in pulmonary nodules (PNs). Chest computed tomography and clinical data of PNs in patients with a history of cancer who underwent surgical resection in The First Affiliated Hospital of Chongqing Medical University from January 1, 2018, to November 30, 2021, were retrospectively collected and evaluated based on Lung-RADS. All PNs were divided into 2 groups: the prior lung cancer (PLC) and the prior extrapulmonary cancer (PEPC) groups. Each group was divided into the ≥5 years and <5 years groups based on the duration of cancer history. The diagnostic agreement of Lung-RADS was evaluated based on the pathological diagnosis of nodules after operation. The diagnostic agreement rate (AR) of Lung-RADS and the composition ratios of different types between different groups were calculated and compared. A total of 451 patients with 565 PNs were included in this study. These patients were divided into the PLC group (<5 years: 135 cases, 175 PNs; ≥5 years: 9 cases, 12 PNs) and the PEPC group (<5 years: 219 cases, 278 PNs; ≥5 years: 88 cases, 100 PNs). The diagnostic AR of partial solid nodules (93.0%; 95% CI: 88.7-97.2%) and solid nodules (88.1%; 95% CI: 84.1-92.1%) was close (P=0.13), while both were higher than that of the pure ground-glass nodules (24.0%; 95% CI: 17.5-30.4%; all P values <0.001). Within 5 years, the composition ratio of PNs and the diagnostic AR (PLC: 58.9%, 95% CI: 51.5-66.2%; PEPC: 76.6%, 95% CI: 71.6-81.6%) between the PLC and PEPC groups were all different (all P values <0.001), and the others [composition ratio of PNs & the diagnostic AR: PLC (≥5 years) vs. PEPC (≥5 years); PLC (<5 years) vs. PLC (≥5 years); PEPC (<5 years) vs. PEPC (≥5 years)] were similar (all P values >0.05; range: 0.10-0.93). The length of prior cancer history may affect the diagnostic agreement of Lung-RADS, especially for patients with prior lung cancer within 5 years.

Assessment of artificial intelligence-aided computed tomography in lung cancer screening

BackgroundLung cancer is one of the most common causes of cancer-related deaths in developed and developing countries. Therefore, early detection of lung cancer has a significant impact on lung cancer surveillance. Interpretation of lung CT scans for cancer screening is considered an intensive task for most radiologists, and long experience is required for accurate diagnosis through visual processing. This cross-sectional study introduces automated CAD software (Careline Soft’s AVIEW Metric software). This software can detect and classify lung nodules in CT scans. The performance of a deep learning (DL) model embedded in that software will be compared with that of the radiologists. Also, the feasibility of lung cancer screening protocol is evaluated in Suez Canal University Hospital, Ismailia, Egypt, by implementing Lung Imaging Reporting and Data System (Lung-RADS).ResultsAs for the detection of the pulmonary nodules, the initial review by the CAD system (without validation by the researcher radiologist) has high sensitivity (93.0%) and specificity (95.5%) with overall accuracy of 93.6%. After review of the automatically detected nodules by the researcher radiologist was done, the final CAD has higher sensitivity (98.2%) and comparable specificity (95.5%) for the detection of pulmonary nodules with overall accuracy of 97.4%. As for lung cancer screening (categorization of Lung-RADS 3 and 4 nodules), unrevised initial computer-aided detection has 97.9% specificity and 96.9% for lung cancer screening with overall accuracy of 97.4%. After second look and review of the CAD result by the researcher radiologist, there is total agreement in total number of nodules and categorization of Lung-RADS 3 and 4. This gives an excellent agreement of 88.6% (κ = 0.951) between the CAD system and reference radiologist in the overall categorization of all lung nodules according to Lung-RADS classification.ConclusionsThe application of CAD system demonstrated increased sensitivity and specificity for the detection of lung nodules and total agreement in the detection of suspicious and probably benign nodules (lung cancer screening) and excellent level of agreement in the overall lung nodule categorization (Lung-RADS).

MRI Compared with Low-Dose CT for Incidental Lung Nodule Detection in COPD: A Multicenter Trial.

To investigate morphofunctional chest MRI for the detection and management of incidental pulmonary nodules in participants with chronic obstructive pulmonary disease (COPD). In this prospective study, 567 participants (mean age, 66 years ± 9 [SD]; 340 men) underwent same-day contrast-enhanced MRI and nonenhanced low-dose CT (LDCT) in a nationwide multicenter trial (clinicaltrials.gov: NCT01245933). Nodule dimensions, morphologic features, and Lung Imaging Reporting and Data System (Lung-RADS) category were assessed at MRI by two blinded radiologists, and consensual LDCT results served as the reference standard. Comparisons were performed using the Student t test, and agreements were assessed using the Cohen weighted κ. A total of 525 nodules larger than 3 mm in diameter were detected at LDCT in 178 participants, with a mean diameter of 7.2 mm ± 6.1 (range, 3.1-63.1 mm). Nodules were not detected in the remaining 389 participants. Sensitivity and positive predictive values with MRI for readers 1 and 2, respectively, were 63.0% and 84.8% and 60.2% and 83.9% for solid nodules (n = 495), 17.6% and 75.0% and 17.6% and 60.0% for part-solid nodules (n = 17), and 7.7% and 100% and 7.7% and 50.0% for ground-glass nodules (n = 13). For nodules 6 mm or greater in diameter, sensitivity and positive predictive values were 73.3% and 92.2% for reader 1 and 71.4% and 93.2% for reader 2, respectively. Readers underestimated the long-axis diameter at MRI by 0.5 mm ± 1.7 (reader 1) and 0.5 mm ± 1.5 (reader 2) compared with LDCT (P < .001). For Lung-RADS categorization per nodule using MRI, there was substantial to perfect interreader agreement (κ = 0.75-1.00) and intermethod agreement compared with LDCT (κ = 0.70-1.00 and 0.69-1.00). In a multicenter setting, morphofunctional MRI showed moderate sensitivity for detection of incidental pulmonary nodules in participants with COPD but high agreement with LDCT for Lung-RADS classification of nodules.Clinical trial registration no. NCT01245933 and NCT02629432Keywords: MRI, CT, Thorax, Lung, Chronic Obstructive Pulmonary Disease, Screening© RSNA, 2023 Supplemental material is available for this article.

Outcomes From More Than 1 Million People Screened for Lung Cancer With Low-Dose CT Imaging

Lung cancer screening (LCS) with low-dose CT (LDCT) imaging was recommended in 2013, making approximately 8 million Americans eligible for LCS. The demographic characteristics and outcomes of individuals screened in the United States have not been reported at the population level. What are the outcomes among people screened and entered in the American College of Radiology's Lung Cancer Screening Registry compared with those of trial participants? This was a cohort study of individuals undergoing baseline LDCT imaging for LCS between 201 and 2019. Predictors of adherence to annual screening were computed. LDCT scan interpretations by Lung Imaging Reporting and Data System (Lung-RADS) score, cancer detection rates (CDRs), and stage at diagnosis were compared with National Lung Cancer Screening Trial (NLST) data. Adherence was 22.3%, and predictors of poor adherence included current smoking status and Hispanic or Black race. On baseline screening, 83%of patients showed negative results and 17%showed positive screening results. The overall CDR was 0.56%. The percentage of people with cancer detected at baseline was higher in the positive Lung-RADS categories at 0.4%for Lung-RADS category 3, 2.6%for Lung-RADS category 4A, 11.1%for Lung-RADS category 4B, and 19.9%for Lung-RADS category 4X. The cancer stage distribution was similar to that observed in the NLST, with 53.5%of patients receiving a diagnosis of stage I cancer and 14.3%with stage IV cancer. Underreporting into the registry may have occurred. This study revealed both the positive aspects of CT scan screening for lung cancer and the challenges that remain. Findings on CT imaging were correlated accurately with lung cancer detection using the Lung-RADS system. A significant stage shift toward early-stage lung cancer was present. Adherence to LCS was poor and likely contributes to the lower than expected cancer detection rate, all of which will impact the outcomes of patients undergoing screening for lung cancer.

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.

Sensitivity and specificity of magnetic resonance imaging in routine diagnosis of pulmonary lesions: a comparison with computed tomography.

State-of-the-art thoracic magnetic resonance imaging (MRI) plays a complementary role in the assessment of pulmonary nodules/masses which potentially indicate to cancer. We aimed to evaluate the sensitivity and specificity of MRI in diagnosis of pulmonary nodules/masses. Sixty-eight patients with computed tomography (CT)-detected pulmonary nodules/masses underwent 3T MRI (T1-VIBE, T1-starVIBE, T2-fBLADE turbo spin-echo, and T2-SPACE). The detection rate was calculated for each of the different subgroups of pulmonary nodules according to lung imaging reporting and data system (Lung-RADS). The four MRI sequences were compared in terms of detection rate and image quality-signal to noise ratio (SNR), contrast to noise ratio (CNR) and 5-point scoring scale. Agreement of lesion size measurement between CT and MRI was assessed by intraclass correlation coefficient (ICC). The picture-SNR, lesion-SNR and CNR of each sequence were analyzed by Mann-Whitney U test. In total, 232 pulmonary lesions were detected by CT. The CT showed 86 solid nodules (SNs) <6 mm, 15 SNs between 6-8 mm, 35 SNs between 8-15 mm, and 52 SNs between 15-30 mm. The T1-VIBE, T1-starVIBE, T2-fBLADE TSE and T2-SPACE sequences accurately detected 141 SNs (141/188, 75%/83.3%), 150 SNs (150/188, 79.8%/100%), 166 SNs (166/188, 88.3%/66.7%) and 169 SNs (169/188, 89.9%/53.3%), respectively. Four ground glass nodules (GGNs) (4/6) were detected by T2-fBLADE TSE. Twelve part-solid nodules (PSNs) (12/22) were detected by T1-VIBE and 20 PSNs (20/22) by T2-SPACE. A total of 100 lesions (2.2±1.4 cm, 0.8-7.3 cm) were accurately detected and measured by the four MRI sequences with ICC >0.96. The picture-SNR, lesion-SNR and CNR by T1-starVIBE were higher than those by T1-VIBE (P<0.001). The lesion-SNR and CNR by T2-fBLADE TSE were higher than those by T2-SPACE (P=0.006, 0.038). 86% of images by T1-starVIBE, 92% by T2-fBLADE TSE, 90% by T2-SPACE and 93% by T1-VIBE were scored 3 or more. MRI achieves high sensitivity and specificity for different type of pulmonary nodules detection and is an effective alternative to CT as a diagnostic tool for pulmonary nodules.

Racial Disparities in Adherence to Annual Lung Cancer Screening and Recommended Follow-Up Care: A Multicenter Cohort Study.

Rationale: Black patients receive recommended lung cancer screening (LCS) follow-up care less frequently than White patients, but it is unknown if this racial disparity persists across both decentralized and centralized LCS programs. Objectives: To determine adherence to American College of Radiology Lung Imaging Reporting and Data System (Lung-RADS) recommendations among individuals undergoing LCS at either decentralized or centralized programs and to evaluate the association of race with LCS adherence. Methods: We performed a multicenter retrospective cohort study of patients receiving LCS at five heterogeneous U.S. healthcare systems. We calculated adherence to annual LCS among patients with a negative baseline screen (Lung-RADS 1 or 2) and recommended follow-up care among those with a positive baseline screen (Lung-RADS 3, 4A, 4B, or 4X) stratified by type of LCS program and evaluated the association between race and adherence using multivariable modified Poisson regression. Results: Of the 6,134 total individuals receiving LCS, 5,142 (83.8%) had negative baseline screens, and 992 (16.2%) had positive baseline screens. Adherence to both annual LCS (34.8% vs. 76.1%; P < 0.001) and recommended follow-up care (63.9% vs. 74.6%; P < 0.001) was lower at decentralized compared with centralized programs. Among individuals with negative baseline screens, a racial disparity in adherence was observed only at decentralized screening programs (interaction term, P < 0.001). At decentralized programs, Black race was associated with 27% reduced adherence to annual LCS (adjusted relative risk [aRR], 0.73; 95% confidence interval [CI], 0.63-0.84), whereas at centralized programs, no effect by race was observed (aRR, 0.98; 95% CI, 0.91-1.05). In contrast, among those with positive baseline screens, there was no significant difference by race for adherence to recommended follow-up care by type of LCS program (decentralized aRR, 0.95; 95% CI, 0.81-1.11; centralized aRR, 0.81; 95% CI, 0.71-0.93; interaction term, P = 0.176). Conclusions: In this large multicenter study of individuals screened for lung cancer, adherence to both annual LCS and recommended follow-up care was greater at centralized screening programs. Black patients were less likely to receive annual LCS than White patients at decentralized compared with centralized LCS programs. Our results highlight the need for further study of healthcare system-level mechanisms to optimize longitudinal LCS care.

Lung Cancer Screening by Race and Ethnicity in an Integrated Health System in Hawaii

Racial and ethnic differences in lung cancer screening (LCS) completion and follow-up may be associated with lung cancer incidence and mortality rates among high-risk populations. Aggregation of Asian American, Native Hawaiian, and Pacific Islander racial and ethnic groups may mask the true underlying disparities in screening uptake and diagnostic follow-up, creating barriers for targeted, preventive health care. To examine racial and ethnic differences in LCS completion and follow-up rates in a multiethnic population. This population-based cohort study was conducted at a health maintenance organization in Hawaii. LCS program participants were identified using electronic medical records from January 1, 2015, to December 31, 2019. Study eligibility requirements included being aged 55 to 79 years, a 30 pack-year smoking history, a current smoker or having quit within the past 15 years, at least 5 years past any lung cancer diagnosis and treatment, and cancer free. Data analysis was performed from June 2019 to October 2020. Eligible for LCS. Screening rates were analyzed by self-reported race and ethnicity and completion of a low-dose computed tomography (LDCT) test. Diagnostic follow-up results were based on the Lung Imaging Reporting and Data System (Lung-RADS) staging system. A total of 1030 eligible LCS program members had an order placed; their mean (SD) age was 65.5 (5.8) years, and 633 (61%) were men. The largest racial and ethnic groups were non-Hispanic White (381 participants [37.0%]), Native Hawaiian or part Native Hawaiian (186 participants [18.1%]), and Japanese (146 participants [14.2%]). Men and Filipino, Chinese, Japanese, and non-Hispanic White individuals had a higher proportion of screen orders for LDCT compared with women and individuals of the other racial and ethnic groups. The overall LCS completion rate was 81% (838 participants). There was a 14% to 15% screening completion rate gap among groups. Asian individuals had the highest screening completion rate (266 participants [86%]) followed by Native Hawaiian (149 participants [80%]) and non-Hispanic White individuals (305 participants [80%]), Pacific Islander (50 participants [79%]) individuals, and individuals of other racial and ethnic groups (68 participants [77%]). Within Asian subgroups, Korean (31 participants [94%]) and Japanese (129 participants [88%]) individuals had the highest completion rates followed by Chinese individuals (28 participants [82%]) and Filipino individuals (78 participants [79%]). Of the 54 participants with Lung-RADS stage 3 disease, 93% (50 participants) completed a 6-month surveillance LDCT test; of 37 individuals with Lung-RADS stage 4 disease, 35 (97%) were followed-up for additional procedures. This cohort study found racial and ethnic disparities in LCS completion rates after disaggregation of Native Hawaiian, Pacific Islander, and Asian individuals and their subgroups. These findings suggest that future research is needed to understand factors that may be associated with LCS completion and follow-up behaviors among these racial and ethnic groups.

Patient Identification of Lung Cancer Screening Follow-Up Recommendations and the Association with Adherence.

Rationale: Adherence to follow-up lung cancer screening (LCS) in real-world settings is suboptimal. Patient understanding of screening results and anticipated follow-up may be crucial to adherence. Objectives: To determine patient factors associated with identification of follow-up recommendations as a measure of patient understanding of screening results after LCS, and to determine whether misidentification of follow-up is associated with lower adherence to recommendations. Methods: We performed a prospective study of patients in the University of Washington/Seattle Cancer Care Alliance LCS registry who underwent an initial LCS examination between June 2017 and September 2019. We mailed potential participants a survey after the initial LCS examination, with additional data abstracted from the electronic health record and LCS registry. Participants were asked to identify the timing and next step for their follow-up, with answers corresponding to the lung imaging reporting and data system (Lung-RADS) recommendations. We examined associations between incorrect identification of recommended follow-up and patient-level characteristics, self-perceived benefit/harm of LCS, LCS knowledge, Lung-RADS score, and patient-reported method of LCS results communication (letter, telephone, or in-person). We used multivariable logistic regression to evaluate associations with incorrect identification of recommendations and assessed incorrect identification of recommendations as a potential mechanism for poor adherence in a separate regression model. Results: One hundred eighty-eight participants completed the survey (response rate 44%); 47% misidentified their follow-up recommendation. Those with Lung-RADS scores ⩾3 had higher odds of incorrectly identifying follow-up recommendations than those with scores <3, as did those with lower educational attainment. However, there was no significant association between incorrect identification of follow-up and ultimate adherence to follow-up. Conclusions: Understanding of LCS follow-up appears to be poor, especially among those with lower education levels and positive findings. Among survey responders, incorrect identification of follow-up was not associated with poor adherence, suggesting that other factors, such as provider interventions, may be driving adherence behavior. These results can inform efforts to target improved patient education regarding follow-up for LCS.

Assisted versus Manual Interpretation of Low-Dose CT Scans for Lung Cancer Screening: Impact on Lung-RADS Agreement.

Purpose To compare the inter- and intraobserver agreement and reading times achieved when assigning Lung Imaging Reporting and Data System (Lung-RADS) categories to baseline and follow-up lung cancer screening studies by using a dedicated CT lung screening viewer with integrated nodule detection and volumetric support with those achieved by using a standard picture archiving and communication system (PACS)-like viewer. Materials and Methods Data were obtained from the National Lung Screening Trial (NLST). By using data recorded by NLST radiologists, scans were assigned to Lung-RADS categories. For each Lung-RADS category (1 or 2, 3, 4A, and 4B), 40 CT scans (20 baseline scans and 20 follow-up scans) were randomly selected for 160 participants (median age, 61 years; interquartile range, 58-66 years; 61 women) in total. Seven blinded observers independently read all CT scans twice in a randomized order with a 2-week washout period: once by using the standard PACS-like viewer and once by using the dedicated viewer. Observers were asked to assign a Lung-RADS category to each scan and indicate the risk-dominant nodule. Inter- and intraobserver agreement was analyzed by using Fleiss κ values and Cohen weighted κ values, respectively. Reading times were compared by using a Wilcoxon signed rank test. Results The interobserver agreement was moderate for the standard viewer and substantial for the dedicated viewer, with Fleiss κ values of 0.58 (95% CI: 0.55, 0.60) and 0.66 (95% CI: 0.64, 0.68), respectively. The intraobserver agreement was substantial, with a mean Cohen weighted κ value of 0.67. The median reading time was significantly reduced from 160 seconds with the standard viewer to 86 seconds with the dedicated viewer (P < .001). Conclusion Lung-RADS interobserver agreement increased from moderate to substantial when using the dedicated CT lung screening viewer. The median reading time was substantially reduced when scans were read by using the dedicated CT lung screening viewer. Keywords: CT, Thorax, Lung, Computer Applications-Detection/Diagnosis, Observer Performance, Technology Assessment Supplemental material is available for this article. © RSNA, 2021.

Application of computer-aided diagnosis for Lung-RADS categorization in CT screening for lung cancer: effect on inter-reader agreement.

To evaluate the effects of computer-aided diagnosis (CAD) on inter-reader agreement in Lung Imaging Reporting and Data System (Lung-RADS) categorization. Two hundred baseline CT scans covering all Lung-RADS categories were randomly selected from the National Lung Cancer Screening Trial. Five radiologists independently reviewed the CT scans and assigned Lung-RADS categories without CAD and with CAD. The CAD system presented up to five of the most risk-dominant nodules with measurements and predicted Lung-RADS category. Inter-reader agreement was analyzed using multirater Fleiss κ statistics. The five readers reported 139-151 negative screening results without CAD and 126-142 with CAD. With CAD, readers tended to upstage (average, 12.3%) rather than downstage Lung-RADS category (average, 4.4%). Inter-reader agreement of five readers for Lung-RADS categorization was moderate (Fleiss kappa, 0.60 [95% confidence interval, 0.57, 0.63]) without CAD, and slightly improved to substantial (Fleiss kappa, 0.65 [95% CI, 0.63, 0.68]) with CAD. The major cause for disagreement was assignment of different risk-dominant nodules in the reading sessions without and with CAD (54.2% [201/371] vs. 63.6% [232/365]). The proportion of disagreement in nodule size measurement was reduced from 5.1% (102/2000) to 3.1% (62/2000) with the use of CAD (p < 0.001). In 31 cancer-positive cases, substantial management discrepancies (category 1/2 vs. 4A/B) between reader pairs decreased with application of CAD (pooled sensitivity, 85.2% vs. 91.6%; p = 0.004). Application of CAD demonstrated a minor improvement in inter-reader agreement of Lung-RADS category, while showing the potential to reduce measurement variability and substantial management change in cancer-positive cases. • Inter-reader agreement of five readers for Lung-RADS categorization was minimally improved by application of CAD, with a Fleiss kappa value of 0.60 to 0.65. • The major cause for disagreement was assignment of different risk-dominant nodules in the reading sessions without and with CAD (54.2% vs. 63.6%). • In 31 cancer-positive cases, substantial management discrepancies between reader pairs, referring to a difference in follow-up interval of at least 9 months (category 1/2 vs. 4A/B), were reduced in half by application of CAD (32/310 to 16/310) (pooled sensitivity, 85.2% vs. 91.6%; p = 0.004).

Feasibility of implementing a national lung cancer screening program: Interim results from the Korean Lung Cancer Screening Project (K-LUCAS).

BackgroundLung cancer screening conducted in high-risk group using low-dose computer tomography (LDCT) has been reported as an effective method to reduce lung cancer mortality in two large randomized-control trials. However, the effectiveness is uncertain when lung cancer screening is expanded to a nationwide population-based program.MethodsThe Korean Lung Cancer Screening Project (K-LUCAS) is a single-arm cohort study that was conducted from February 2017 to evaluate the feasibility of implementing an organized national lung cancer screening program in Korea. High-risk population aged 55–74 years with more than a 30-pack-year smoking history was recruited. Smoking history was obtained from administering questionnaires at national health screening programs or public smoking cessation programs which are already established programs in Korea. The screening results were reported using the Lung Imaging Reporting and Data System (Lung-RADS), suggested by the American College of Radiology. K-LUCAS was performed by a network-based diagnosis supporting system using a computer-aided detection (CAD) program to maintain screening quality. Current smokers were provided with mandatory smoking counseling.ResultsAmong 71,829 participants aged 50 years or older in the national health screening program, 5,975 (8.3%) were eligible for lung cancer screening. Among them, 1,062 (17.8%) refused to participate in K-LUCAS. Additionally, 779 participants were recruited in the smoking cessation program. Thus, a total of 5,692 eligible high-risk participants were recruited in this study. Among them, 865 (15.2%) had positive screening results, which requires a further examination; 529 (9.3%) had Lung-RADS category 3 (indeterminate), and 336 (5.9%) had category 4 (suspicious of lung cancer); 42 (0.7%) had confirmed lung cancer. Approximately 66.7% had early-stage lung cancer: 24 (57.1%), stage I and 4 (9.5%), stage II. Six (1.1%) patients developed complications at the time of diagnosis, including one death. The anxiety level related to cancer screening was low. Participation in screening encouraged motivation to quit smoking.ConclusionsK-LUCAS provided promising evidence supporting the implementation of a national lung cancer screening program to detect early stage lung cancer and promote smoking cessation for participants in Asian population.

Distribution of Lung-RADS categories according to job type in a single shipyard workers.

BackgroundRecently, lung cancer screenings based on age and smoking history using low-dose computed tomography (LDCT) have begun in Korea. This study aimed to evaluate the distribution of lung imaging reporting and data system (Lung-RADS) categories in shipyard workers exposed to lung carcinogens such as nickel, chromium, and welding fumes according to job type, to provide basic data regarding indications for LDCT in shipyard workers.MethodsThis study included 6,326 workers from a single shipyard, who underwent health examinations with LDCT between January 2010 and December 2018. Data on age, smoking status and history, medical history, and job type were investigated. The participants were categorized into high-exposure, low-exposure, and non-exposure job groups based on the estimated exposure level of nickel, chromium, and welding fumes according to job type. Cox proportional hazard regression analysis was used to determine the difference between exposure groups in Lung-RADS category ≥ 3 (3, 4A, and 4B).ResultsOut of all participants, 97 (1.5%) participants were classified into Lung-RADS category ≥ 3 and 7 (0.1%) participants were confirmed as lung cancer. The positive predictive value (ratio of diagnosed lung cancer cases to Lung-RADS category ≥ 3) was 7.2%. The hazard ratio (HR) of Lung-RADS category ≥ 3 was 1.451 (95% confidence interval [CI]: 0.911–2.309) in low-exposure and 1.692 (95% CI: 1.007–2.843) in high-exposure job group. Adjusting for age and pack-years, the HR was statistically significant only in the high-exposure job group (HR: 1.689; 95% CI: 1.004–2.841).ConclusionsBased on LDCT and Lung-RADS, among male shipyard workers, Lung-RADS category ≥ 3 were significantly higher in the high-exposure job group. Their HR tended to be > 1.0 and was statistically significant in the high-exposure job group. Additional studies should be conducted to establish more elaborate LDCT indications for occupational health examination.

Combination of Deep Learning-Based Denoising and Iterative Reconstruction for Ultra-Low-Dose CT of the Chest: Image Quality and Lung-RADS Evaluation.

OBJECTIVE. The objective of our study was to assess the effect of the combination of deep learning-based denoising (DLD) and iterative reconstruction (IR) on image quality and Lung Imaging Reporting and Data System (Lung-RADS) evaluation on chest ultra-low-dose CT (ULDCT). MATERIALS AND METHODS. Forty-one patients with 252 nodules were evaluated retrospectively. All patients underwent ULDCT (mean ± SD, 0.19 ± 0.01 mSv) and standard-dose CT (SDCT) (6.46 ± 2.28 mSv). ULDCT images were reconstructed using hybrid iterative reconstruction (HIR) and model-based iterative reconstruction (MBIR), and they were postprocessed using DLD (i.e., HIR-DLD and MBIR-DLD). SDCT images were reconstructed using filtered back projection. Three independent radiologists subjectively evaluated HIR, HIR-DLD, MBIR, and MBIR-DLD images on a 5-point scale in terms of noise, streak artifact, nodule edge, clarity of small vessels, homogeneity of the normal lung parenchyma, and overall image quality. Two radiologists independently evaluated the nodules according to Lung-RADS using HIR, MBIR, HIR-DLD, and MBIR-DLD ULDCT images and SDCT images. The median scores for subjective analysis were analyzed using Wilcoxon signed rank test with Bonferroni correction. Intraobserver agreement for Lung-RADS category between ULDCT and SDCT was evaluated using the weighted kappa coefficient. RESULTS. In the subjective analysis, ULDCT with DLD showed significantly better scores than did ULDCT without DLD (p < 0.001), and MBIR-DLD showed the best scores among the ULDCT images (p < 0.001) for all items. In the Lung-RADS evaluation, HIR showed fair or moderate agreement (reader 1 and reader 2: κw = 0.46 and 0.32, respectively); MBIR, moderate or good agreement (κw = 0.68 and 0.57); HIR-DLD, moderate agreement (κw = 0.53 and 0.48); and MBIR-DLD, good agreement (κw = 0.70 and 0.72). CONCLUSION. DLD improved the image quality of both HIR and MBIR on ULDCT. MBIR-DLD was superior to HIR_DLD for image quality and for Lung-RADS evaluation.