Multifactorial Analysis of Mortality in Screening Detected Lung Cancer

AI Body Composition in Lung Cancer Screening: Added Value Beyond Lung Cancer Detection

Background An artificial intelligence (AI) algorithm has been developed for fully automated body composition assessment of lung cancer screening noncontrast low-dose CT of the chest (LDCT) scans, but the utility of these measurements in disease risk prediction models has not been assessed. Purpose To evaluate the added value of CT-based AI-derived body composition measurements in risk prediction of lung cancer incidence, lung cancer death, cardiovascular disease (CVD) death, and all-cause mortality in the National Lung Screening Trial (NLST). Materials and Methods In this secondary analysis of the NLST, body composition measurements, including area and attenuation attributes of skeletal muscle and subcutaneous adipose tissue, were derived from baseline LDCT examinations by using a previously developed AI algorithm. The added value of these measurements was assessed with sex- and cause-specific Cox proportional hazards models with and without the AI-derived body composition measurements for predicting lung cancer incidence, lung cancer death, CVD death, and all-cause mortality. Models were adjusted for confounding variables including age; body mass index; quantitative emphysema; coronary artery calcification; history of diabetes, heart disease, hypertension, and stroke; and other PLCOM2012 lung cancer risk factors. Goodness-of-fit improvements were assessed with the likelihood ratio test. Results Among 20 768 included participants (median age, 61 years [IQR, 57-65 years]; 12 317 men), 865 were diagnosed with lung cancer and 4180 died during follow-up. Including the AI-derived body composition measurements improved risk prediction for lung cancer death (male participants: χ2 = 23.09, P < .001; female participants: χ2 = 15.04, P = .002), CVD death (males: χ2 = 69.94, P < .001; females: χ2 = 16.60, P < .001), and all-cause mortality (males: χ2 = 248.13, P < .001; females: χ2 = 94.54, P < .001), but not for lung cancer incidence (male participants: χ2 = 2.53, P = .11; female participants: χ2 = 1.73, P = .19). Conclusion The body composition measurements automatically derived from baseline low-dose CT examinations added predictive value for lung cancer death, CVD death, and all-cause death, but not for lung cancer incidence in the NLST. Clinical trial registration no. NCT00047385 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Fintelmann in this issue.

Artificial intelligence-based vessel suppression for detection of sub-solid nodules in lung cancer screening computed tomography

Background

Lung cancer screening (LCS) with low-dose computed tomography (LDCT) helps early lung cancer detection, commonly presenting as small pulmonary nodules. Artificial intelligence (AI)-based vessel suppression (AI-VS) and automatic detection (AI-AD) algorithm can improve detection of subsolid nodules (SSNs) on LDCT. We assessed the impact of AI-VS and AI-AD in detection and classification of SSNs [ground-glass nodules (GGNs) and part-solid nodules (PSNs)], on LDCT performed for LCS.

Methods

Following regulatory approval, 123 LDCT examinations with sub-solid pulmonary nodules (average diameter ≥6 mm) were processed to generate three image series for each examination-unprocessed, AI-VS, and AI-AD series with annotated lung nodules. Two thoracic radiologists in consensus formed the standard of reference (SOR) for this study. Two other thoracic radiologists (R1 and R2; 5 and 10 years of experience in thoracic CT image interpretation) independently assessed the unprocessed images alone, then together with AI-VS series, and finally with AI-AD for detecting all ≥6 mm GGN and PSN. We performed receiver operator characteristics (ROC) and Cohen's Kappa analyses for statistical analyses.

Results

On unprocessed images, R1 and R2 detected 232/310 nodules (R1: 114 GGN, 118 PSN) and 255/310 nodules (R2: 122 GGN, 133 PSN), respectively (P>0.05). On AI-VS images, they detected 249/310 nodules (119 GGN, 130 PSN) and 277/310 nodules (128 GGN, 149 PSN), respectively (P≥0.12). When compared to the SOR, accuracy (AUC) for detection of PSN on the AI-VS images (AUC 0.80-0.81) was greater than on the unprocessed images (AUC 0.70-0.76). AI-VS images enabled detection of solid components in five nodules deemed as GGN on the unprocessed images. Accuracy of AI-AD was lower than both the radiologists (AUC 0.60-0.72).

Conclusions

AI-VS improved the detection and classification of SSN into GGN and PSN on LDCT of the chest for the two radiologist (R1 and R2) readers.

Management and Outcomes of Suspected Infectious and Inflammatory Lung Abnormalities Identified on Lung Cancer Screening CT

Background: Incidental findings are frequently encountered during lung cancer screening (LCS). Limited data describe the prevalence of suspected acute infectious and inflammatory lung processes on LCS and how they should be managed. Objective: To determine the prevalence, radiologic reporting and management, and outcome of suspected infectious and inflammatory lung processes identified incidentally during LCS, and to propose a management algorithm. Methods: This retrospective study included 6314 low dose CT (LDCT) examinations performed between June 2014 and April 2019 in 3800 patients as part of an established LCS program. Radiology reports were reviewed, and patients with potentially infectious or inflammatory lung abnormalities were identified and analyzed for descriptors of imaging findings, Lung-RADS designation, recommendations, and clinical outcomes. Based on the descriptors, outcomes and a >2% threshold risk of malignancy, a follow-up algorithm was developed to decrease additional imaging without affecting cancer detection. Results: A total of 331/3800 (8.7%) patients (178 men, 153 women; mean age: 66 ± 7 years) undergoing LCS had lung findings that were attributed to infection or inflammation. These abnormalities were reported as potentially significant findings using the "S" modifier in 149/331 (45.0%) and as the "dominant nodule" determining the Lung-RADS category in 96/331 (29.0%). Abnormalities were multiple or multifocal in 260/331 (78.5%). Common descriptors were ground-glass (155/331; 46.8%), tree-in-bud (56/331; 16.9%), consolidation (41/331; 12.4%), and clustered (67/331; 20.2%) opacities. A follow-up chest CT outside of screening was performed within 12 months or less in 264/331 (79.8%) and within 6 months or less in 286/331 (56.2%). A total of 260/331 (78.5%) opacities resolved on follow-up imaging. Two malignancies (2/331; 0.60%) were associated with these abnormalities, and both had consolidations. Theoretical adoption of a proposed management algorithm for suspected infectious and inflammatory findings reduced unnecessary follow-up imaging by 82.6% without missing a single malignancy. Conclusions: Presumed acute infectious or inflammatory lung abnormalities are frequently encountered in the setting of LCS. These opacities are commonly multifocal and resolve on follow-up. Less than 1% are associated with malignancy. Clinical impact: Adoption of a conservative management algorithm can standardize recommendations and reduce unnecessary imaging without increasing the risk of missing a malignancy.

Quantitative Pectoralis Muscle Area is Associated with the Development of Lung Cancer in a Large Lung Cancer Screening Cohort

BACKGROUND

Studies have demonstrated an inverse relationship between body mass index (BMI) and the risk of developing lung cancer. We conducted a retrospective cohort study evaluating baseline quantitative computed tomography (CT) measurements of body composition, specifically muscle and fat area in a large CT lung screening cohort (CTLS). We hypothesized that quantitative measurements of baseline body composition may aid in risk stratification for lung cancer.

METHODS

Patients who underwent baseline CTLS between January 1st, 2012 and September 30th, 2014 and who had an in-network primary care physician were included. All patients met NCCN Guidelines eligibility criteria for CTLS. Quantitative measurements of pectoralis muscle area (PMA) and subcutaneous fat area (SFA) were performed on a single axial slice of the CT above the aortic arch with the Chest Imaging Platform Workstation software. Cox multivariable proportional hazards model for cancer was adjusted for variables with a univariate p < 0.2. Data were dichotomized by sex and then combined to account for baseline differences between sexes.

RESULTS

One thousand six hundred and ninety six patients were included in this study. A total of 79 (4.7%) patients developed lung cancer. There was an association between the 25th percentile of PMA and the development of lung cancer [HR 1.71 (1.07, 2.75), p < 0.025] after adjusting for age, BMI, qualitative emphysema, qualitative coronary artery calcification, and baseline Lung-RADS® score.

CONCLUSIONS

Quantitative assessment of PMA on baseline CTLS was associated with the development of lung cancer. Quantitative PMA has the potential to be incorporated as a variable in future lung cancer risk models.

Emphysema - The review of radiological presentation and its clinical impact in the LDCT screening era

Emphysema is one of three main lung pathologies in Chronic Obstructive Pulmonary Disease, along with chronic bronchitis and small airway obstruction. The diagnosis is based on detection of low attenuation areas in lung tissue on chest Computed Tomography, either visual by a radiologist, or automatic by the applied Computed Tomography software. Results of the studies on the association between emphysema and lung cancer incidence are mixed. Many studies have demonstrated, that chronic lung diseases, like Chronic Obstructive Pulmonary Disease, are associated with lung cancer morbidity. There is also evidence, that emphysema can be related with worse prognosis in patients with detected lung cancer. In this review article we aim to summarize current knowledge about emphysema detection and evaluation on Computed Tomography, both quantitative and qualitative. We also summarize current data on correlation between emphysema and lung cancer, as well as its potential use in selecting patients, who would most benefit from lung cancer screening.

The association of silicosis severity with pectoralis major muscle and subcutaneous fat volumes and the pulmonary artery/aorta ratio evaluated by CT

PURPOSE

Silicosis is an incurable occupational disease that sometimes rapidly progresses with fatal outcomes. We aimed to evaluate the association between disease severity and the change in the pectoralis major muscle volume (PMV), subcutaneous fat volume (SFV), and the pulmonary artery/aorta (P/Ao) ratio in patients with silicosis using computed tomography (CT).

METHODS

The study included 41 male silicosis patients and 41 control group subjects with available chest CT images. Using dedicated software, we measured PMV and SFV from the axial CT images. We calculated the P/Ao ratio and obtained body mass index (BMI) and forced expiratory volume/forced vital capacity (FEV1/FVC) results from hospital records. We used the chest X-ray profusion score according to the International Labor Organization (ILO) classification to evaluate the severity of the silicosis.

RESULTS

The mean age was 33.5±4.4 and 34.7±4.7 years in the silicotic and control groups, respectively. The mean BMI, PMV, SFV, and P/Ao values significantly differed between the study and control groups (P = 0.0009, P < 0.0001, P < 0.0001, and P = 0.0029, respectively). According to the ILO classification, there were 12 silicosis patients in category 1, 13 in category 2, and 16 in category 3. A significant difference was found between disease categories in terms of PMV, SFV, P/Ao, BMI, and FEV1/FVC values (P = 0.0425, P = 0.0341, P = 0.0002, P = 0.0492, and P = 0.0004, respectively).

CONCLUSION

Disease severity had a stronger association with decreased PMV and SFV and increased P/Ao ratios than BMI in patients with silicosis caused by denim sandblasting. Thus, CT evaluation might be a useful indicator of disease severity.

Knowledge-Based Analysis for Mortality Prediction From CT Images

Low-Dose CT (LDCT) can significantly improve the accuracy of lung cancer diagnosis and thus reduce cancer deaths compared to chest X-ray. The lung cancer risk population is also at high risk of other deadly diseases, for instance, cardiovascular diseases. Therefore, predicting the all-cause mortality risks of this population is of great importance. This paper introduces a knowledge-based analytical method using deep convolutional neural network (CNN) for all-cause mortality prediction. The underlying approach combines structural image features extracted from CNNs, based on LDCT volume at different scales, and clinical knowledge obtained from quantitative measurements, to predict the mortality risk of lung cancer screening subjects. The proposed method is referred as Knowledge-based Analysis of Mortality Prediction Network (KAMP-Net). It constitutes a collaborative framework that utilizes both imaging features and anatomical information, instead of completely relying on automatic feature extraction. Our work demonstrates the feasibility of incorporating quantitative clinical measurements to assist CNNs in all-cause mortality prediction from chest LDCT images. The results of this study confirm that radiologist defined features can complement CNNs in performance improvement. The experiments demonstrate that KAMP-Net can achieve a superior performance when compared to other methods.

Hybrid Neural Networks for Mortality Prediction from LDCT Images

Known for its high morbidity and mortality rates, lung cancer poses a significant threat to human health and well-being. However, the same population is also at high risk for other deadly diseases, such as cardiovascular disease. Since Low-Dose CT (LDCT) has been shown to significantly improve the lung cancer diagnosis accuracy, it will be very useful for clinical practice to predict the all-cause mortality for lung cancer patients to take corresponding actions. In this paper, we propose a deep learning based method, which takes both chest LDCT image patches and coronary artery calcification risk scores as input to predict the mortality risk of lung cancer subjects. The proposed method is called Hybrid Risk Network (HyRiskNet) for mortality risk prediction, which is an end-to-end framework utilizing hybrid imaging features, instead of completely relying on automatic feature extraction. Our work demonstrates the feasibility of using deep learning techniques for all-cause lung cancer mortality prediction from chest LDCT images. The experimental results show that HyRiskNet can achieve superior performance compared with the neural networks with only image input and with other traditional semi-automatic scoring methods. The study also indicates that radiologist defined features can well complement convolutional neural networks for more comprehensive feature extraction.

Can CT radiomic analysis in NSCLC predict histology and EGFR mutation status?

To assess the role of radiomic features in distinguishing squamous and adenocarcinoma subtypes of nonsmall cell lung cancers (NSCLC) and predict EGFR mutations.Institution Review Board-approved study included chest CT scans of 93 consecutive patients (43 men, 50 women, mean age 60 ± 11 years) with biopsy-proven squamous and adenocarcinoma lung cancers greater than 1 cm. All cancers were evaluated for epidermal growth factor receptor (EGFR) mutation. The clinical parameters such as age, sex, and smoking history and standard morphology-based CT imaging features such as target lesion longest diameter (LD), longest perpendicular diameter (LPD), density, and presence of cavity were recorded. The radiomics data was obtained using commercial CT texture analysis (CTTA) software. The CTTA was performed on a single image of the dominant lung lesion. The predictive value of clinical history, standard imaging features, and radiomics was assessed with multivariable logistic regression and receiver operating characteristic (ROC) analyses.Between adenocarcinoma and squamous cell carcinomas, ROC analysis showed significant difference in 3/11 radiomic features (entropy, normalized SD, total) [AUC 0.686-0.744, P = .006 to <.0001], 1/3 clinical features (smoking) [AUC 0.732, P = .001], and 2/3 imaging features (LD and LPD) [AUC 0.646-0658, P = .020 to .032]. ROC analysis for probability variables showed higher values for radiomics (AUC 0.800, P < .0001) than clinical (AUC 0.676, P = .017) and standard imaging (AUC 0.708, P < .0001). Between EGFR mutant and wild-type adenocarcinoma, ROC analysis showed significant difference in 2/11 radiomic features (kurtosis, K2) [AUC 0.656-0.713, P = .03 to .003], 1/3 clinical features (smoking) [AUC 0.758, P < .0001]. The combined probability variable for radiomics, clinical and imaging features was higher (AUC 0.890, P < .0001) than independent probability variables.The radiomics evaluation adds incremental value to clinical history and standard imaging features in predicting histology and EGFR mutations.