ABSTRACT
Low muscle mass is associated with numerous adverse outcomes independent of other associated comorbid diseases. We aimed to predict and understand an individual's risk for developing low muscle mass using proteomics and machine learning. We identified eight biomarkers associated with low pectoralis muscle area (PMA). We built three random forest classification models that used either clinical measures, feature selected biomarkers, or both to predict development of low PMA. The area under the receiver operating characteristic curve for each model was: clinical-only = 0.646, biomarker-only = 0.740, and combined = 0.744. We displayed the heterogenetic nature of an individual's risk for developing low PMA and identified two distinct subtypes of participants who developed low PMA. While additional validation is required, our methods for identifying and understanding individual and group risk for low muscle mass could be used to enable developments in the personalized prevention of low muscle mass.
Subject(s)
Biomarkers , Machine Learning , Pectoralis Muscles , Proteomics , Humans , Proteomics/methods , Male , Female , Adult , Middle Aged , ROC CurveABSTRACT
BACKGROUND AND OBJECTIVES: The scarcity of data for training deep learning models in pediatrics has prompted questions about the feasibility of employing CNNs trained with adult images for pediatric populations. In this work, a pneumonia classification CNN was used as an exploratory example to showcase the adaptability and efficacy of such models in pediatric healthcare settings despite the inherent data constraints. METHODS: To develop a curated training dataset with reduced biases, 46,947 chest X-ray images from various adult datasets were meticulously selected. Two preprocessing approaches were tried to assess the impact of thoracic segmentation on model attention outside the thoracic area. Evaluation of our approach was carried out on a dataset containing 5,856 chest X-rays of children from 1 to 5 years old. RESULTS: An analysis of attention maps indicated that networks trained with thorax segmentation placed less attention on regions outside the thorax, thus eliminating potential bias. The ensuing network exhibited impressive performance when evaluated on an adult dataset, achieving a pneumonia discrimination AUC of 0.95. When tested on a pediatric dataset, the pneumonia discrimination AUC reached 0.82. CONCLUSIONS: The results of this study show that adult-trained CNNs can be effectively applied to pediatric populations. This could potentially shift focus towards validating adult models over pediatric population instead of training new CNNs with limited pediatric data. To ensure the generalizability of deep learning models, it is important to implement techniques aimed at minimizing biases, such as image segmentation or low-quality image exclusion.
Subject(s)
Pneumonia , Humans , Pneumonia/diagnostic imaging , Adult , Child, Preschool , Infant , Deep Learning , Neural Networks, Computer , Pediatrics/education , Radiography, Thoracic , Male , Female , ChildABSTRACT
Background: The global coronavirus disease 2019 (COVID-19) pandemic has posed substantial challenges for healthcare systems, notably the increased demand for chest computed tomography (CT) scans, which lack automated analysis. Our study addresses this by utilizing artificial intelligence-supported automated computer analysis to investigate lung involvement distribution and extent in COVID-19 patients. Additionally, we explore the association between lung involvement and intensive care unit (ICU) admission, while also comparing computer analysis performance with expert radiologists' assessments. Methods: A total of 81 patients from an open-source COVID database with confirmed COVID-19 infection were included in the study. Three patients were excluded. Lung involvement was assessed in 78 patients using CT scans, and the extent of infiltration and collapse was quantified across various lung lobes and regions. The associations between lung involvement and ICU admission were analysed. Additionally, the computer analysis of COVID-19 involvement was compared against a human rating provided by radiological experts. Results: The results showed a higher degree of infiltration and collapse in the lower lobes compared to the upper lobes (P<0.05). No significant difference was detected in the COVID-19-related involvement of the left and right lower lobes. The right middle lobe demonstrated lower involvement compared to the right lower lobes (P<0.05). When examining the regions, significantly more COVID-19 involvement was found when comparing the posterior vs. the anterior halves and the lower vs. the upper half of the lungs. Patients, who required ICU admission during their treatment exhibited significantly higher COVID-19 involvement in their lung parenchyma according to computer analysis, compared to patients who remained in general wards. Patients with more than 40% COVID-19 involvement were almost exclusively treated in intensive care. A high correlation was observed between computer detection of COVID-19 affections and the rating by radiological experts. Conclusions: The findings suggest that the extent of lung involvement, particularly in the lower lobes, dorsal lungs, and lower half of the lungs, may be associated with the need for ICU admission in patients with COVID-19. Computer analysis showed a high correlation with expert rating, highlighting its potential utility in clinical settings for assessing lung involvement. This information may help guide clinical decision-making and resource allocation during ongoing or future pandemics. Further studies with larger sample sizes are warranted to validate these findings.
ABSTRACT
Low muscle mass is associated with numerous adverse outcomes independent of other associated comorbid diseases. We aimed to predict and understand an individual's risk for developing low muscle mass using proteomics and machine learning. We identified 8 biomarkers associated with low pectoralis muscle area (PMA). We built 3 random forest classification models that used either clinical measures, feature selected biomarkers, or both to predict development of low PMA. The area under the receiver operating characteristic curve for each model was: clinical-only = 0.646, biomarker-only = 0.740, and combined = 0.744. We displayed the heterogenetic nature of an individual's risk for developing low PMA and identified 2 distinct subtypes of participants who developed low PMA. While additional validation is required, our methods for identifying and understanding individual and group risk for low muscle mass could be used to enable developments in the personalized prevention of low muscle mass.
ABSTRACT
Background: Chronic obstructive pulmonary disease (COPD) is a significant public health concern and intercepting the development of emphysema is vital for COPD prevention. Smokers are a high-risk population for emphysema with limited prevention strategies. We aimed to determine if adherence to a nutritionally rich, plant-centered diet among young ever-smokers is associated with reduced risk of future radiographic emphysema. Methods: We studied participants from the Coronary Artery Risk Development in Young Adults (CARDIA) Lung Prospective Cohort Study who were 18-30 years old at enrollment and followed for 30 years. We analyzed 1706 adults who reported current or former smoking by year 20. Repeated measures of diet history were used to calculate A Priori Diet Quality Scores (APDQSs), and categorized into quintiles, with higher quintiles representing higher nutritionally rich plant-centered food intake. Emphysema was assessed at year 25 (n=1351) by computed tomography (CT). Critical covariates were selected, acknowledging potential residual confounding. Results: Emphysema was observed in 13.0% of the cohort, with a mean age of 50.4 ± 3.5 years. The prevalence of emphysema was 4.5% in the highest APDQS quintile (nutritionally rich), compared with 25.4% in the lowest quintile. After adjustment for multiple covariates, including smoking, greater adherence to a plant-centered diet was inversely associated with emphysema (highest versus lowest quintile odds ratio: 0.44, 95% CI 0.19-0.99, ptrend=0.008). Conclusion: Longitudinal adherence to a nutritionally rich, plant-centered diet was associated with a decreased risk of emphysema development in middle adulthood, warranting further examination of diet as a strategy for emphysema prevention in a high-risk smoking population.
ABSTRACT
BACKGROUND: Early recognition of cardiac dysfunction in patients with chronic obstructive pulmonary disease (COPD) may prevent future cardiac impairment and improve prognosis. Quantitative assessment of subsegmental and segmental vessel volume by Computed Tomographic (CT) imaging can provide a surrogate of pulmonary vascular remodeling. We aimed to examine the relationship between lung segmental- and subsegmental vessel volume, and echocardiographic measures of cardiac structure and function in patients with COPD. METHODS: We studied 205 participants with COPD, included in a large cohort study of cardiovascular disease in COPD patients. Participants had an available CT scan and echocardiogram. Artificial intelligence (AI) algorithms calculated the subsegmental vessel fraction as the vascular volume in vessels below 10 mm2 in cross-sectional area, indexed to total intrapulmonary vessel volume. Linear regressions were conducted, and standardized ß-coefficients were calculated. Scatterplots were created to visualize the continuous correlations between the vessel fractions and echocardiographic parameters. RESULTS: We found that lower subsegmental vessel fraction and higher segmental vessel volume were correlated with higher left ventricular (LV) mass, LV diastolic dysfunction, and inferior vena cava (IVC) dilatation. Subsegmental vessel fraction was correlated with right ventricular (RV) remodeling, while segmental vessel fraction was correlated with higher pulmonary pressure. Measures of LV mass and right atrial pressure displayed the strongest correlations with pulmonary vasculature measures. CONCLUSION: Pulmonary vascular remodeling in patients with COPD, may negatively affect cardiac structure and function. AI-identified remodeling in pulmonary vasculature may provide a tool for early identification of COPD patients at higher risk for cardiac impairment.
Subject(s)
Pulmonary Disease, Chronic Obstructive , Vascular Remodeling , Humans , Cohort Studies , Artificial Intelligence , Predictive Value of Tests , Pulmonary Disease, Chronic Obstructive/complications , Pulmonary Disease, Chronic Obstructive/diagnostic imagingABSTRACT
Lung cancer screening involves the use of thoracic CT for both detection and measurements of suspicious lung nodules to guide the screening management. Since lung cancer screening eligibility typically requires age over 50 years along with >20 pack-year tobacco exposure, thoracic CT scans also frequently reveal evidence for pulmonary emphysema as well as coronary artery calcification. These three thoracic diseases are collectively three of the leading causes of premature death across the world. Screening for the major thoracic diseases in this heavily tobacco-exposed cohort is broadening the focus of lung cancer screening to a more comprehensive health evaluation including discussing the relevance of screen-detected findings of the heart and lung parenchyma. The status and implications of these emerging issues were reviewed in a multidisciplinary workshop focused on the process of quantitative imaging in the lung cancer screening setting to guide the evolution of this important new area of public health.
Subject(s)
Lung Neoplasms , Thoracic Diseases , Humans , Middle Aged , Lung Neoplasms/epidemiology , Early Detection of Cancer/methods , Tomography, X-Ray Computed/methods , LungABSTRACT
OBJECTIVES: There have been limited investigations of the prevalence and mortality impact of quantitative computed tomography (QCT) parenchymal lung features in rheumatoid arthritis (RA). We examined the cross-sectional prevalence and mortality associations of QCT features, comparing RA and non-RA participants. METHODS: We identified participants with and without RA in COPDGene, a multicentre cohort study of current or former smokers. Using a k-nearest neighbor quantifier, high resolution CT chest scans were scored for percentage of normal lung, interstitial changes, and emphysema. We examined associations between QCT features and RA using multivariable linear regression. After dichotomizing participants at the 75th percentile for each QCT feature among non-RA participants, we investigated mortality associations by RA/non-RA status and quartile 4 vs quartiles 1-3 of QCT features using Cox regression. We assessed for statistical interactions between RA and QCT features. RESULTS: We identified 82 RA cases and 8820 non-RA comparators. In multivariable linear regression, RA was associated with higher percentage of interstitial changes (ß = 1.7 ± 0.5, p= 0.0008) but not emphysema (ß = 1.3 ± 1.7, p= 0.44). Participants with RA and >75th percentile of emphysema had significantly higher mortality than non-RA participants (HR 5.86, 95%CI 3.75-9.13) as well as RA participants (HR 5.56, 95%CI 2.71-11.38) with ≤75th percentile of emphysema. There were statistical interactions between RA and emphysema for mortality (multiplicative p= 0.014; attributable proportion 0.53, 95%CI 0.30-0.70). CONCLUSIONS: Using machine learning-derived QCT data in a cohort of smokers, RA was associated with higher percentage of interstitial changes. The combination of RA and emphysema conferred >5-fold higher mortality.
ABSTRACT
Pulmonary arterial hypertension associated with schistosomiasis (SchPAH) and pulmonary arterial hypertension associated with portal hypertension (PoPAH) are lung diseases that develop in the presence of liver diseases. However, mechanistic pathways by which the underlying liver conditions and other drivers contribute to the development and progression of pulmonary arterial hypertension (PAH) are unclear for both etiologies. In turn, these unknowns limit certainty of strategies to prevent, diagnose, and reverse the resultant PAH. Here we consider specific mechanisms that contribute to SchPAH and PoPAH, identifying those that may be shared and those that appear to be unique to each etiology, in the hope that this exploration will both highlight known causal drivers and identify knowledge gaps appropriate for future research. Overall, the key pathophysiologic differences that we identify between SchPAH and PoPAH suggest that they are not variants of a single condition.
ABSTRACT
Toll-like receptor 7 (TLR7) is known for eliciting immunity against single-stranded RNA viruses, and is increased in both human and cigarette smoke (CS)-induced, experimental chronic obstructive pulmonary disease (COPD). Here we show that the severity of CS-induced emphysema and COPD is reduced in TLR7-deficient mice, while inhalation of imiquimod, a TLR7-agonist, induces emphysema without CS exposure. This imiquimod-induced emphysema is reduced in mice deficient in mast cell protease-6, or when wild-type mice are treated with the mast cell stabilizer, cromolyn. Furthermore, therapeutic treatment with anti-TLR7 monoclonal antibody suppresses CS-induced emphysema, experimental COPD and accumulation of pulmonary mast cells in mice. Lastly, TLR7 mRNA is increased in pre-existing datasets from patients with COPD, while TLR7+ mast cells are increased in COPD lungs and associated with severity of COPD. Our results thus support roles for TLR7 in mediating emphysema and COPD through mast cell activity, and may implicate TLR7 as a potential therapeutic target.
Subject(s)
Emphysema , Pulmonary Disease, Chronic Obstructive , Pulmonary Emphysema , Humans , Animals , Mice , Tryptases/genetics , Toll-Like Receptor 7/genetics , Imiquimod , Lung , Pulmonary Emphysema/genetics , Nicotiana , Mice, Inbred C57BLABSTRACT
BACKGROUND: Airway mucus plugs are frequently identified on CT scans of patients with COPD with a smoking history without mucus-related symptoms (ie, cough, phlegm [silent mucus plugs]). RESEARCH QUESTION: In patients with COPD, what are the risk and protective factors associated with silent airway mucus plugs? Are silent mucus plugs associated with functional, structural, and clinical measures of disease? STUDY DESIGN AND METHODS: We identified mucus plugs on chest CT scans of participants with COPD from the COPDGene study. The mucus plug score was defined as the number of pulmonary segments with mucus plugs, ranging from 0 to 18, and categorized into three groups (0, 1-2, and ≥ 3). We determined risk and protective factors for silent mucus plugs and the associations of silent mucus plugs with measures of disease severity using multivariable linear and logistic regression models. RESULTS: Of 4,363 participants with COPD, 1,739 had no cough or phlegm. Among the 1,739 participants, 627 (36%) had airway mucus plugs identified on CT scan. Risk factors of silent mucus plugs (compared with symptomatic mucus plugs) were older age (OR, 1.02), female sex (OR, 1.40), and Black race (OR, 1.93) (all P values < .01). Among those without cough or phlegm, silent mucus plugs (vs absence of mucus plugs) were associated with worse 6-min walk distance, worse resting arterial oxygen saturation, worse FEV1 % predicted, greater emphysema, thicker airway walls, and higher odds of severe exacerbation in the past year in adjusted models. INTERPRETATION: Mucus plugs are common in patients with COPD without mucus-related symptoms. Silent mucus plugs are associated with worse functional, structural, and clinical measures of disease. CT scan-identified mucus plugs can complement the evaluation of patients with COPD.
ABSTRACT
Open international challenges are becoming the de facto standard for assessing computer vision and image analysis algorithms. In recent years, new methods have extended the reach of pulmonary airway segmentation that is closer to the limit of image resolution. Since EXACT'09 pulmonary airway segmentation, limited effort has been directed to the quantitative comparison of newly emerged algorithms driven by the maturity of deep learning based approaches and extensive clinical efforts for resolving finer details of distal airways for early intervention of pulmonary diseases. Thus far, public annotated datasets are extremely limited, hindering the development of data-driven methods and detailed performance evaluation of new algorithms. To provide a benchmark for the medical imaging community, we organized the Multi-site, Multi-domain Airway Tree Modeling (ATM'22), which was held as an official challenge event during the MICCAI 2022 conference. ATM'22 provides large-scale CT scans with detailed pulmonary airway annotation, including 500 CT scans (300 for training, 50 for validation, and 150 for testing). The dataset was collected from different sites and it further included a portion of noisy COVID-19 CTs with ground-glass opacity and consolidation. Twenty-three teams participated in the entire phase of the challenge and the algorithms for the top ten teams are reviewed in this paper. Both quantitative and qualitative results revealed that deep learning models embedded with the topological continuity enhancement achieved superior performance in general. ATM'22 challenge holds as an open-call design, the training data and the gold standard evaluation are available upon successful registration via its homepage (https://atm22.grand-challenge.org/).
Subject(s)
Lung Diseases , Trees , Humans , Tomography, X-Ray Computed/methods , Image Processing, Computer-Assisted/methods , Algorithms , Lung/diagnostic imagingABSTRACT
Quantitative assessment of emphysema in CT scans has mostly focused on calculating the percentage of lung tissue that is deemed abnormal based on a density thresholding strategy. However, this overall measure of disease burden discards virtually all the spatial information encoded in the scan that is implicitly utilized in a visual assessment. This simplification is likely grouping heterogenous disease patterns and is potentially obscuring clinical phenotypes and variable disease outcomes. To overcome this, several methods that attempt to quantify heterogeneity in emphysema distribution have been proposed. Here, we compare three of those: one based on estimating a power law for the size distribution of contiguous emphysema clusters, a second that looks at the number of emphysema-to-emphysema voxel adjacencies, and a third that applies a parametric spatial point process model to the emphysema voxel locations. This was done using data from 587 individuals from Phase 1 of COPDGene that had an inspiratory CT scan and plasma protein abundance measurements. The associations between these imaging metrics and visual assessment with clinical measures (FEV[Formula: see text], FEV[Formula: see text]-FVC ratio, etc.) and plasma protein biomarker levels were evaluated using a variety of regression models. Our results showed that a selection of spatial measures had the ability to discern heterogeneous patterns among CTs that had similar emphysema burdens. The most informative quantitative measure, average cluster size from the point process model, showed much stronger associations with nearly every clinical outcome examined than existing CT-derived emphysema metrics and visual assessment. Moreover, approximately 75% more plasma biomarkers were found to be associated with an emphysema heterogeneity phenotype when accounting for spatial clustering measures than when they were excluded.
Subject(s)
Emphysema , Pulmonary Emphysema , Humans , Pulmonary Emphysema/diagnostic imaging , Emphysema/diagnostic imaging , Benchmarking , Lung/diagnostic imaging , Cluster AnalysisABSTRACT
Background: The aim of the current study was to investigate the distribution and extent of lung involvement in patients with COVID-19 with AI-supported, automated computer analysis and to assess the relationship between lung involvement and the need for intensive care unit (ICU) admission. A secondary aim was to compare the performance of computer analysis with the judgment of radiological experts. Methods: A total of 81 patients from an open-source COVID database with confirmed COVID-19 infection were included in the study. Three patients were excluded. Lung involvement was assessed in 78 patients using computed tomography (CT) scans, and the extent of infiltration and collapse was quantified across various lung lobes and regions. The associations between lung involvement and ICU admission were analyzed. Additionally, the computer analysis of COVID-19 involvement was compared against a human rating provided by radiological experts. Results: The results showed a higher degree of infiltration and collapse in the lower lobes compared to the upper lobes (p < 0.05) No significant difference was detected in the COVID-19-related involvement of the left and right lower lobes. The right middle lobe demonstrated lower involvement compared to the right lower lobes (p < 0.05). When examining the regions, significantly more COVID-19 involvement was found when comparing the posterior vs. the anterior halves of the lungs and the lower vs. the upper half of the lungs. Patients, who required ICU admission during their treatment exhibited significantly higher COVID-19 involvement in their lung parenchyma according to computer analysis, compared to patients who remained in general wards. Patients with more than 40% COVID-19 involvement were almost exclusively treated in intensive care. A high correlation was observed between computer detection of COVID-19 affections and expert rating by radiological experts. Conclusion: The findings suggest that the extent of lung involvement, particularly in the lower lobes, dorsal lungs, and lower half of the lungs, may be associated with the need for ICU admission in patients with COVID-19. Computer analysis showed a high correlation with expert rating, highlighting its potential utility in clinical settings for assessing lung involvement. This information may help guide clinical decision-making and resource allocation during ongoing or future pandemics. Further studies with larger sample sizes are warranted to validate these findings.
ABSTRACT
The factors associated with persistent hypoxemia after pulmonary embolus (PE) are not well understood. Predicting the need for oxygen post discharge at the time of diagnosis using available CT imaging will enable better discharge planning. To examine the relationship between CT derived imaging markers (automated computation of arterial small vessel fraction, pulmonary artery diameter to aortic diameter ratio (PA:A), right to left ventricular diameter ratio (RV:LV) and new oxygen requirement at the time of discharge in patients diagnosed with acute intermediate-risk PE. CT measurements were obtained in a retrospective cohort of patients with acute-intermediate risk PE admitted to Brigham and Women's Hospital between 2009 and 2017. Twenty one patients without a history of lung disease requiring home oxygen and 682 patients without discharge oxygen requirements were identified. There was an increased median PA:A ratio (0.98 vs. 0.92, p = 0.02) and arterial small vessel fraction (0.32 vs. 0.39, p = 0.001) in the oxygen-requiring group], but no difference in the median RV:LV ratio (1.20 vs. 1.20, p = 0.74). Being in the upper quantile for the arterial small vessel fraction was associated with decreased odds of oxygen requirement (OR 0.30 [0.10-0.78], p = 0.02). Loss of arterial small vessel volume as measured by arterial small vessel fraction and an increase in the PA:A ratio at the time of diagnosis were associated with the presence of persistent hypoxemia on discharge in acute intermediate-risk PE.
Subject(s)
Pulmonary Embolism , Ventricular Dysfunction, Right , Humans , Female , Tomography, X-Ray Computed/methods , Retrospective Studies , Aftercare , Predictive Value of Tests , Patient Discharge , Hypoxia , Oxygen , Acute DiseaseABSTRACT
BACKGROUND: Sarcopenia, or loss of skeletal muscle mass and decreased contractile strength, contributes to morbidity and mortality in patients with chronic obstructive pulmonary disease (COPD). The severity of sarcopenia in COPD is variable, and there are limited data to explain phenotype heterogeneity. Others have shown that COPD patients with sarcopenia have several hallmarks of cellular senescence, a potential mechanism of primary (age-related) sarcopenia. We tested if genetic contributors explain the variability in sarcopenic phenotype and accelerated senescence in COPD. METHODS: To identify gene variants [single nucleotide polymorphisms (SNPs)] associated with sarcopenia in COPD, we performed a genome-wide association study (GWAS) of fat free mass index (FFMI) in 32 426 non-Hispanic White (NHW) UK Biobank participants with COPD. Several SNPs within the fat mass and obesity-associated (FTO) gene were associated with sarcopenia that were validated in an independent COPDGene cohort (n = 3656). Leucocyte telomere length quantified in the UK Biobank cohort was used as a marker of senescence. Experimental validation was done by genetic depletion of FTO in murine skeletal myotubes exposed to prolonged intermittent hypoxia or chronic hypoxia because hypoxia contributes to sarcopenia in COPD. Molecular biomarkers for senescence were also quantified with FTO depletion in murine myotubes. RESULTS: Multiple SNPs located in the FTO gene were associated with sarcopenia in addition to novel SNPs both within and in proximity to the gene AC090771.2, which transcribes long non-coding RNA (lncRNA). To replicate our findings, we performed a GWAS of FFMI in NHW subjects from COPDGene. The SNP most significantly associated with FFMI was on chromosome (chr) 16, rs1558902A > T in the FTO gene (ß = 0.151, SE = 0.021, P = 1.40 × 10-12 for UK Biobank |ß= 0.220, SE = 0.041, P = 9.99 × 10-8 for COPDGene) and chr 18 SNP rs11664369C > T nearest to the AC090771.2 gene (ß = 0.129, SE = 0.024, P = 4.64 × 10-8 for UK Biobank |ß = 0.203, SE = 0.045, P = 6.38 × 10-6 for COPDGene). Lower handgrip strength, a measure of muscle strength, but not FFMI was associated with reduced telomere length in the UK Biobank. Experimentally, in vitro knockdown of FTO lowered myotube diameter and induced a senescence-associated molecular phenotype, which was worsened by prolonged intermittent hypoxia and chronic hypoxia. CONCLUSIONS: Genetic polymorphisms of FTO and AC090771.2 were associated with sarcopenia in COPD in independent cohorts. Knockdown of FTO in murine myotubes caused a molecular phenotype consistent with senescence that was exacerbated by hypoxia, a common condition in COPD. Genetic variation may interact with hypoxia and contribute to variable severity of sarcopenia and skeletal muscle molecular senescence phenotype in COPD.
Subject(s)
Pulmonary Disease, Chronic Obstructive , Sarcopenia , Animals , Mice , Sarcopenia/genetics , Sarcopenia/complications , Hand Strength , Genome-Wide Association Study , Pulmonary Disease, Chronic Obstructive/genetics , Pulmonary Disease, Chronic Obstructive/complications , Polymorphism, Single Nucleotide , HypoxiaABSTRACT
BACKGROUND: Radiomics, defined as quantitative features extracted from images, provide a non-invasive means of assessing malignant versus benign pulmonary nodules. In this study, we evaluate the consistency with which perinodular radiomics extracted from low-dose computed tomography images serve to identify malignant pulmonary nodules. MATERIALS AND METHODS: Using the National Lung Screening Trial (NLST), we selected individuals with pulmonary nodules between 4mm to 20mm in diameter. Nodules were segmented to generate four distinct datasets; 1) a Tumor dataset containing tumor-specific features, 2) a 10 mm Band dataset containing parenchymal features between the segmented nodule boundary and 10mm out from the boundary, 3) a 15mm Band dataset, and 4) a Tumor Size dataset containing the maximum nodule diameter. Models to predict malignancy were constructed using support-vector machine (SVM), random forest (RF), and least absolute shrinkage and selection operator (LASSO) approaches. Ten-fold cross validation with 10 repetitions per fold was used to evaluate the performance of each approach applied to each dataset. RESULTS: With respect to the RF, the Tumor, 10mm Band, and 15mm Band datasets achieved areas under the receiver-operator curve (AUC) of 84.44%, 84.09%, and 81.57%, respectively. Significant differences in performance were observed between the Tumor and 15mm Band datasets (adj. p-value <0.001). However, when combining tumor-specific features with perinodular features, the 10mm Band + Tumor and 15mm Band + Tumor datasets (AUC 87.87% and 86.75%, respectively) performed significantly better than the Tumor Size dataset (66.76%) or the Tumor dataset. Similarly, the AUCs from the SVM and LASSO were 84.71% and 88.91%, respectively, for the 10mm Band + Tumor. CONCLUSIONS: The combined 10mm Band + Tumor dataset improved the differentiation between benign and malignant lung nodules compared to the Tumor datasets across all methodologies. This demonstrates that parenchymal features capture novel diagnostic information beyond that present in the nodule itself. (data agreement: NLST-163).
Subject(s)
Adenocarcinoma of Lung , Lung Neoplasms , Multiple Pulmonary Nodules , Humans , Lung Neoplasms/diagnostic imaging , Lung Neoplasms/pathology , Lung/pathology , Adenocarcinoma of Lung/pathology , Multiple Pulmonary Nodules/pathology , Tomography, X-Ray Computed/methods , Retrospective StudiesABSTRACT
BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a disease of accelerated aging and is associated with comorbid conditions including osteoporosis and sarcopenia. These extrapulmonary conditions are highly prevalent yet frequently underdiagnosed and overlooked by pulmonologists in COPD treatment and management. There is evidence supporting a role for bone-muscle crosstalk which may compound osteoporosis and sarcopenia risk in COPD. Chest CT is commonly utilized in COPD management, and we evaluated its utility to identify low bone mineral density (BMD) and reduced pectoralis muscle area (PMA) as surrogates for osteoporosis and sarcopenia. We then tested whether BMD and PMA were associated with morbidity and mortality in COPD. METHODS: BMD and PMA were analyzed from chest CT scans of 8468 COPDGene participants with COPD and controls (smoking and non-smoking). Multivariable regression models tested the relationship of BMD and PMA with measures of function (6-min walk distance (6MWD), handgrip strength) and disease severity (percent emphysema and lung function). Multivariable Cox proportional hazards models were used to evaluate the relationship between sex-specific quartiles of BMD and/or PMA derived from non-smoking controls with all-cause mortality. RESULTS: COPD subjects had significantly lower BMD and PMA compared with controls. Higher BMD and PMA were associated with increased physical function and less disease severity. Participants with the highest BMD and PMA quartiles had a significantly reduced mortality risk (36% and 46%) compared to the lowest quartiles. CONCLUSIONS: These findings highlight the potential for CT-derived BMD and PMA to characterize osteoporosis and sarcopenia using equipment available in the pulmonary setting.
Subject(s)
Osteoporosis , Pulmonary Disease, Chronic Obstructive , Sarcopenia , Humans , Male , Female , Sarcopenia/diagnostic imaging , Sarcopenia/epidemiology , Hand Strength , Osteoporosis/diagnostic imaging , Osteoporosis/epidemiology , Osteoporosis/complications , Tomography, X-Ray Computed/adverse effects , Morbidity , Muscles , Bone DensityABSTRACT
The emerging field of radiomics that transforms standard-of-care images to quantifiable scalar statistics endeavors to reveal the information hidden in these macroscopic images. The concept of texture is widely used and essential in many radiomic-based studies. Practice usually reduces spatial multidimensional texture matrices, e.g., gray-level co-occurrence matrices (GLCMs), to summary scalar features. These statistical features have been demonstrated to be strongly correlated and tend to contribute redundant information; and does not account for the spatial information hidden in the multivariate texture matrices. This study proposes a novel pipeline to deal with spatial texture features in radiomic studies. A new set of textural features that preserve the spatial information inherent in GLCMs is proposed and used for classification purposes. The set of the new features uses the Wasserstein metric from optimal mass transport theory (OMT) to quantify the spatial similarity between samples within a given label class. In particular, based on a selected subset of texture GLCMs from the training cohort, we propose new representative spatial texture features, which we incorporate into a supervised image classification pipeline. The pipeline relies on the support vector machine (SVM) algorithm along with Bayesian optimization and the Wasserstein metric. The selection of the best GLCM references is considered for each classification label and is performed during the training phase of the SVM classifier using a Bayesian optimizer. We assume that sample fitness is defined based on closeness (in the sense of the Wasserstein metric) and high correlation (Spearman's rank sense) with other samples in the same class. Moreover, the newly defined spatial texture features consist of the Wasserstein distance between the optimally selected references and the remaining samples. We assessed the performance of the proposed classification pipeline in diagnosing the coronavirus disease 2019 (COVID-19) from computed tomographic (CT) images. To evaluate the proposed spatial features' added value, we compared the performance of the proposed classification pipeline with other SVM-based classifiers that account for different texture features, namely: statistical features only, optimized spatial features using Euclidean metric, non-optimized spatial features with Wasserstein metric. The proposed technique, which accounts for the optimized spatial texture feature with Wasserstein metric, shows great potential in classifying new COVID CT images that the algorithm has not seen in the training step. The MATLAB code of the proposed classification pipeline is made available. It can be used to find the best reference samples in other data cohorts, which can then be employed to build different prediction models.