Deep Learning Classification of Usual Interstitial Pneumonia Predicts Outcomes.

Rationale: Computed tomography (CT) enables noninvasive diagnosis of usual interstitial pneumonia (UIP), but enhanced image analyses are needed to overcome the limitations of visual assessment. Objectives: Apply multiple instance learning (MIL) to develop an explainable deep learning algorithm for prediction of UIP from CT and validate its performance in independent cohorts. Methods: We trained an MIL algorithm using a pooled dataset (n = 2,143) and tested it in three independent populations: data from a prior publication (n = 127), a single-institution clinical cohort (n = 239), and a national registry of patients with pulmonary fibrosis (n = 979). We tested UIP classification performance using receiver operating characteristic analysis, with histologic UIP as ground truth. Cox proportional hazards and linear mixed-effects models were used to examine associations between MIL predictions and survival or longitudinal FVC. Measurements and Main Results: In two cohorts with biopsy data, MIL improved accuracy for histologic UIP (area under the curve, 0.77 [n = 127] and 0.79 [n = 239]) compared with visual assessment (area under the curve, 0.65 and 0.71). In cohorts with survival data, MIL-UIP classifications were significant for mortality (n = 239, mortality to April 2021: unadjusted hazard ratio, 3.1; 95% confidence interval [CI], 1.96-4.91; P < 0.001; and n = 979, mortality to July 2022: unadjusted hazard ratio, 3.64; 95% CI, 2.66-4.97; P < 0.001). Individuals classified as UIP positive by the algorithm had a significantly greater annual decline in FVC than those classified as UIP negative (-88 ml/yr vs. -45 ml/yr; n = 979; P < 0.01), adjusting for extent of lung fibrosis. Conclusions: Computerized assessment using MIL identifies clinically significant features of UIP on CT. Such a method could improve confidence in radiologic assessment of patients with interstitial lung disease, potentially enabling earlier and more precise diagnosis.

Deep Learning-based Fibrosis Extent on Computed Tomography Predicts Outcome of Fibrosing Interstitial Lung Disease Independent of Visually Assessed Computed Tomography Pattern.

Rationale: Radiologic pattern has been shown to predict survival in patients with fibrosing interstitial lung disease. The additional prognostic value of fibrosis extent by quantitative computed tomography (CT) is unknown. Objectives: We hypothesized that fibrosis extent provides information beyond visually assessed CT pattern that is useful for outcome prediction. Methods: We performed a retrospective analysis of chest CT, demographics, longitudinal pulmonary function, and transplantation-free survival among participants in the Pulmonary Fibrosis Foundation Patient Registry. CT pattern was classified visually according to the 2018 usual interstitial pneumonia criteria. Extent of fibrosis was objectively quantified using data-driven textural analysis. We used Kaplan-Meier plots and Cox proportional hazards and linear mixed-effects models to evaluate the relationships between CT-derived metrics and outcomes. Results: Visual assessment and quantitative analysis were performed on 979 enrollment CT scans. Linear mixed-effect modeling showed that greater baseline fibrosis extent was significantly associated with the annual rate of decline in forced vital capacity. In multivariable models that included CT pattern and fibrosis extent, quantitative fibrosis extent was strongly associated with transplantation-free survival independent of CT pattern (hazard ratio, 1.04; 95% confidence interval, 1.04-1.05; P < 0.001; C statistic = 0.73). Conclusions: The extent of lung fibrosis by quantitative CT is a strong predictor of physiologic progression and survival, independent of visually assessed CT pattern.

Pictorial Review of Fibrotic Interstitial Lung Disease on High-Resolution CT Scan and Updated Classification.

TOPIC IMPORTANCE: Given the recently expanded approval of antifibrotics for various fibrotic interstitial lung diseases (ILDs), early and correct recognition of these diseases is imperative for physicians. Because high-resolution chest CT scan forms the backbone of diagnosis for ILD, this review will discuss evidence-based imaging findings of key fibrotic ILDs and an approach for differentiating these diseases. REVIEW FINDINGS: (1) Imaging findings of nonspecific interstitial pneumonia may evolve over time and become indistinguishable from usual interstitial pneumonia. Therefore, if remote imaging can be reviewed, this would increase the likelihood of an accurate imaging diagnosis, particularly if findings appear to represent a usual interstitial pneumonia pattern on the recent examination. (2) Given the difficulty and lack of objectivity in classifying patients with hypersensitivity pneumonitis into acute, subacute, and chronic categories and that prognosis depends primarily on presence or absence of fibrosis, the new set of guidelines released in 2020 categorizes patients with hypersensitivity pneumonitis as either nonfibrotic (purely inflammatory) or fibrotic (either purely fibrotic or mixed fibrotic/inflammatory) based on imaging and/or histologic findings, and the prior temporal terms are no longer used. (3) Interstitial lung abnormalities are incidental CT scan findings that may suggest early ILD in patients without clinical suspicion for ILD. Patients with high-risk features should undergo clinical evaluation for ILD and be actively monitored for disease progression. SUMMARY: Fibrotic ILD on high-resolution chest CT scan is a complex topic, but with use of an evidence-based analysis and algorithm as provided in this article, the probability of a correct imaging diagnosis increases.

Findings on High Resolution Computed Tomography in Symptomatic Veterans with Deployment-Related Lung Disease.

PURPOSE: Military deployment to dusty, austere environments in Southwest Asia and Afghanistan is associated with symptomatic airways diseases including asthma and bronchiolitis. The utility of chest high-resolution computed tomographic (HRCT) imaging in lung disease diagnosis in this population is poorly understood. We investigated visual assessment of HRCT for identifying deployment-related lung disease compared with healthy controls. MATERIALS AND METHODS: Chest HRCT images from 46 healthy controls and 45 symptomatic deployed military personnel with clinically confirmed asthma and/or biopsy-confirmed distal lung disease were scored by 3 independent thoracic radiologists. We compared demographic and clinical characteristics and frequency of imaging findings between deployers and controls, and between deployers with asthma and those with biopsy-confirmed distal lung disease, using χ2, Fisher exact or t tests, and logistic regression where appropriate. We also analyzed inter-rater agreement for imaging findings. RESULTS: Expiratory air trapping was the only chest CT imaging finding that was significantly more frequent in deployers compared with controls. None of the 24 deployers with biopsy-confirmed bronchiolitis and/or granulomatous pneumonitis had HRCT findings of inspiratory mosaic attenuation or centrilobular nodularity. Only 2 of 21 with biopsy-proven emphysema had emphysema on HRCT. CONCLUSIONS: Compared with surgical lung biopsy, visual assessment of HRCT showed few abnormalities in this small cohort of previously deployed symptomatic veterans with normal or near-normal spirometry.

Interstitial Lung Abnormality-Why Should I Care and What Should I Do About It?

Interstitial lung abnormalities (ILAs) are specific computed tomography (CT) findings that are potentially compatible with interstitial lung disease (ILD) in patients without clinical suspicion for disease. ILAs are associated with adverse clinical outcomes including increased mortality, imaging progression and lung function decline, and increased lung injury risk with lung cancer therapies. It is expected that identification of ILAs will increase with implementation of lung cancer screening and diagnostic CT imaging for workup of other pathologies. As such, radiologists will play a critical role in the diagnosis and management of ILAs.

Emphysema Progression at CT by Deep Learning Predicts Functional Impairment and Mortality: Results from the COPDGene Study.

Background Visual assessment remains the standard for evaluating emphysema at CT; however, it is time consuming, is subjective, requires training, and is affected by variability that may limit sensitivity to longitudinal change. Purpose To evaluate the clinical and imaging significance of increasing emphysema severity as graded by a deep learning algorithm on sequential CT scans in cigarette smokers. Materials and Methods A secondary analysis of the prospective Genetic Epidemiology of Chronic Obstructive Pulmonary Disease (COPDGene) study participants was performed and included baseline and 5-year follow-up CT scans from 2007 to 2017. Emphysema was classified automatically according to the Fleischner emphysema grading system at baseline and 5-year follow-up using a deep learning model. Baseline and change in clinical and imaging parameters at 5-year follow-up were compared in participants whose emphysema progressed versus those who did not. Kaplan-Meier analysis and multivariable Cox regression were used to assess the relationship between emphysema score progression and mortality. Results A total of 5056 participants (mean age, 60 years ± 9 [SD]; 2566 men) were evaluated. At 5-year follow-up, 1293 of the 5056 participants (26%) had emphysema progression according to the Fleischner grading system. This group demonstrated progressive airflow obstruction (forced expiratory volume in 1 second [percent predicted]: -3.4 vs -1.8), a greater decline in 6-minute walk distance (-177 m vs -124 m), and greater progression in quantitative emphysema extent (adjusted lung density: -1.4 g/L vs 0.5 g/L; percentage of lung voxels with CT attenuation less than -950 HU: 0.6 vs 0.2) than those with nonprogressive emphysema (P < .001 for each). Multivariable Cox regression analysis showed a higher mortality rate in the group with emphysema progression, with an estimated hazard ratio of 1.5 (95% CI: 1.2, 1.8; P < .001). Conclusion An increase in Fleischner emphysema grade on sequential CT scans using an automated deep learning algorithm was associated with increased functional impairment and increased risk of mortality. ClinicalTrials.gov registration no. NCT00608764 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Grenier in this issue.

Quantitative imaging analysis detects subtle airway abnormalities in symptomatic military deployers.

BACKGROUND: Exposure to inhalational hazards during post-9/11 deployment to Southwest Asia and Afghanistan puts military personnel at risk for respiratory symptoms and disease. Pulmonary function and qualitative chest high resolution computed tomography (HRCT) are often normal in "deployers" with persistent respiratory symptoms. We explored the utility of quantitative HRCT imaging markers of large and small airways abnormalities, including airway wall thickness, emphysema, and air trapping, in symptomatic deployers with clinically-confirmed lung disease compared to controls. METHODS: Chest HRCT images from 45 healthy controls and 82 symptomatic deployers with asthma, distal lung disease or both were analyzed using Thirona Lung quantification software to calculate airway wall thickness (by Pi10), emphysema (by percentage of lung volume with attenuation < -950 Hounsfield units [LAA%-950]), and three parameters of air trapping (expiratory/inspiratory total lung volume and mean lung density ratios, and LAA%-856). SAS v.9.4 was used to compare demographic and clinical characteristics between deployers and controls using Chi-Square, Fisher Exact or t-tests. Linear regression was used to assess relationships between pulmonary function and quantitative imaging findings. RESULTS: Gender and smoking status were not statistically significantly different between groups, but deployers were significantly younger than controls (42 vs 58 years, p < 0.0001), had higher body mass index (31 vs 28 kg/m2, p = 0.01), and had fewer total smoking pack-years (8 vs. 26, p = 0.007). Spirometric measures were not statistically significantly different between groups. Pi10 and LAA%-950 were significantly elevated in deployers compared to controls in unadjusted analyses, with the emphysema measure remaining significantly higher in deployers after adjustment for age, sex, smoking, BMI, and expiratory total lung volume. Air trapping parameters were more common in control images, likely due to differences in age and smoking between groups. Among deployers, LAA%-950 and Pi10 were significantly correlated with spirometric markers of obstruction based on ratio of forced expiratory volume in one second (FEV1)/forced vital capacity (FVC) and/or percent predicted FEV1. CONCLUSIONS: Quantitative chest HRCT imaging analysis identifies emphysema in deployers with asthma and distal lung disease, and may be useful in detecting and monitoring deployment-related lung disease in a population where spirometry is typically normal.

Imaging of the rare cystic lung diseases.

When discussing cystic lung diseases, a certain group of diseases tends to receive the majority of attention. Other less frequently discussed cystic lung diseases are also important causes of morbidity in patients. Etiologies include genetic syndromes, lymphoproliferative diseases, infections, exogenous exposures, and a developmental abnormality. This review article focuses on the clinical and imaging features of these other cystic lung diseases.

Visual Emphysema at Chest CT in GOLD Stage 0 Cigarette Smokers Predicts Disease Progression: Results from the COPDGene Study.

Background The clinical significance of visually evident emphysema on CT images in individuals without spirometric evidence of chronic obstructive pulmonary disease (COPD) by current diagnostic criteria is, to the knowledge of the authors, unknown. Purpose To evaluate whether participants with visually evident emphysema at CT were more likely to have progressive disease and increased mortality at 5 years compared with those without visual emphysema. Materials and Methods This secondary analysis of the prospective Genetic Epidemiology of COPD study evaluated current or former smokers enrolled between 2008 and 2011 who did not meet current criteria for COPD (defined as Global Initiative for Obstructive Lung Disease stage 0). Statistical analysis was performed by using linear mixed models to estimate mean physiologic, imaging, and clinical outcomes for those with and without visual emphysema. Hazard ratios for mortality were calculated by using Cox regression models with emphysema as the main predictor. Results Of the 4095 participants, 48.3% (1979 participants; 1096 men and 883 women; mean age, 57 years ± 8 [standard deviation]) had trace or greater visual emphysema at CT and 51.7% (2116 participants; 1068 men and 1048 women; mean age, 56 years ± 8) had no emphysema at CT. At 5 years, participants with visual emphysema at CT demonstrated progressive airflow obstruction with lower values of ratio of forced expiratory volume in 1 second (FEV1)-to-functional vital capacity (FVC) ratio (-1.7 vs -0.7) and greater progression in quantitative emphysema measured by 15th percentile lung density (-3.3 vs -0.3 HU), adjusted lung density (-3.1 vs -0.2 g/L), and percentage of lung voxels with CT attenuation less than -950 HU (0.17 vs -0.20) than participants without emphysema (P < .001 for each). The rate of quantitative emphysema progression increased with greater grades of emphysema severity within the emphysema group. Conclusion The presence of visual emphysema at CT in current and former Global Initiative for Obstructive Lung Disease stage 0 smokers predicted structural and physiologic disease progression. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Grenier in this issue.

Imaging-histologic discordance at percutaneous biopsy of the lung.

RATIONALE AND OBJECTIVES: The purpose of this study was to quantify the degree of imaging-histologic discordance in a cohort of patients undergoing computed tomography (CT)-guided lung biopsy for focal lung disease. MATERIALS AND METHODS: A retrospective review was performed of 186 patients who underwent percutaneous lung biopsy of a parenchymal lesion at our institution between January and December 2009. Diagnostic radiology reports of CT or positron emission tomography-CTs performed before biopsy were used to classify the lesion as malignant or benign by five readers. Pathology reports of the biopsied lesions were classified by three readers. Inter-reader agreement and imaging-histologic concordance were quantified using kappa statistics. Discordant benign cases were then revisited to determine downstream effects. RESULTS: Inter-reader agreement on report content was substantial or almost perfect with kappas >0.783. Kappas for concordance were as follows: malignant (0.448), primary lung cancer (0.517), metastatic disease to lung (0.449), benign (0.510), and overall agreement (0.381). Of the twelve discordant benign cases that were revisited, four were found to be false negatives, resulting in a delay in diagnosis. CONCLUSIONS: Our study of imaging-histologic discordance in percutaneous biopsy of lung lesions supports the need for imaging report standardization and improved integration and communication between the fields of radiology and pathology.