Citations

Abstract Introduction: Endobronchial optical coherence tomography (EB-OCT) imaging has unique potential to diagnose and quantify early, microscopic disease changes and assess therapeutic responsivity in interstitial lung disease (ILD). However, training and time requirements for interpretation and quantitative evaluation of large volumetric EB-OCT datasets need to be optimized to facilitate clinical adoption. We develop and validate a computationally-efficient, rapid EB-OCT quantitative image analysis framework using artificial intelligence (AI) with deep learning architecture for 2D feature segmentation, 3D feature burden and spatial distribution mapping, and computer-aided diagnosis in early ILD subjects. Methods: Volumetric EB-OCT datasets, consisting of 2D cross-sectional images, were acquired from multiple locations in the bilateral lungs of ILD and healthy control subjects. EB-OCT features were manually segmented from cross-sectional images by an expert EB-OCT reader, including fibrosis, microscopic honeycombing, traction bronchiolectasis, normal parenchyma, and emphysema. EB-OCT images with corresponding feature segmentations were used to train and validate a multiclass convolutional neural network. The 2D AI-based segmentation model was subsequently evaluated against manual ‘ground truth’ segmentations for each feature in a de novo ‘test’ dataset of independent subjects using dice similarity score, precision, recall/sensitivity, specificity, and balanced accuracy. To generate 3D AI-based quantitative feature burden and spatial distribution maps, AI-based feature segmentation was performed on each EB-OCT 2D cross-sectional image within a volumetric imaging site. Three external pathologist readers independently interpreted 3D AI-based features maps for each EB-OCT imaging site acquired for a subject (without access to EB-OCT images) and provided a single diagnosis of usual interstitial pneumonia (UIP) or non-UIP ILD. Performance of computer-aided ILD diagnosis was compared against independently evaluated histopathology, clinical diagnosis, and prior EB-OCT image interpretation. Results: EB-OCT was acquired in 51 subjects: 30 in vivo early ILD, 5 in vivo non-ILD controls, and 16 ex vivo end-stage ILD. Overall pixel similarity between 2D manual and AI-based segmentation demonstrated high accuracy for each ILD feature (mean balanced accuracy 0.9). 3D quantitative AI-based feature maps demonstrated strong agreement (Spearman ρ0.87, p<0.001) between manual and AI-based segmentation. AI-based computer-aided diagnosis demonstrated 100% sensitivity/specificity for histologic UIP and clinical IPF for all pathologist readers, with improved diagnostic accuracy and significantly reduced training (45-minute training, 75% reduction) and interpretation time (<1 minute per subject for computer-aided diagnosis, 88% reduction) compared to EB-OCT image interpretation. Conclusion: The AI-based EB-OCT image analysis framework could expedite computer-aided diagnosis of early ILD and robust, quantitative assessment of microscopic disease burden, progression, and therapeutic responsivity.

Abstract Background Sarcoidosis, amyloidosis, hemochromatosis and scleroderma are the most common systemic disorders leading to infiltrative cardiomyopathy (InCM). Each of them has been associated with conduction abnormalities and sudden cardiac death. In patients who undergo in-hospital cardiac arrest, a high index of suspicion is required to rule out InCM as an underlying contributor. Purpose We aimed to analyse the temporal trend in the prevalence of InCM among patients with in-hospital cardiac arrest and identify factors associated with increased mortality in this population. Methods We analysed data from the National Inpatient Sample (NIS), which is the largest publicly available database of in-hospital patients in the Unites States. Using ICD-10 coding, we identified patients who were hospitalized across 10 years from 2010–2019. Patients with diagnosis codes for both cardiac arrest and a subtype of InCM were included. Demographic characteristics and co-morbidity data were collected. Statistical analysis was performed using STATA. Results The total number of patients with a diagnosis of in-hospital cardiac arrest was 19,34,260. Of these, the total number with InCM was 14,803 (0.77%), which comprised our study population. Mean age was 63 years. The overall prevalence of InCM across the years ranged between 0.75 to 0.9%, with a significant temporal increase (p<0.01). Prevalence of females ranged between 51% to 62% and significantly outnumbered males across the time period. (p=0.011). Incidence of in-hospital mortality ranged between 61% to 76% for females and 30 to 38% for males. Race based analysis showed that prevalence of Caucasians ranged from 51% to 62% while that of African Americans (AA) ranged from 57% to 73%. The incidence of mortality was similar in both populations. Sarcoidosis was the most prevalent form of InCM, accounting for nearly two-fifths of all cases while amyloidosis and scleroderma both accounted for nearly 20% each. Haemochromatosis was present in 10% of the population. Overall mortality rate in our study population was 60.6%. The following co-morbidities were more prevalent in patients with InCM than those without: history of heart failure, chronic obstructive pulmonary disease (COPD), chronic kidney disease, anaemia, malignancy, coagulopathy, ventricular tachycardia, acute kidney injury and stroke. The following factors were independent predictors of in-hospital mortality-age (p=0.001), female gender (p=0.016), Hispanic race (p=0.039), history of COPD (p<0.001) and presence of malignancy (p=0.042). Conclusion The prevalence of infiltrative cardiomyopathy in patients with in-hospital cardiac arrest is increasing. Females, older patients and Hispanic population are at an increased risk of mortality. Sex and race based disparities in the prevalence of InCM in patients with in-hospital cardiac arrest is an area of further research. Funding Acknowledgement Type of funding sources: None.

Abstract Rationale Early, accurate diagnosis of interstitial lung disease (ILD) informs prognosis and therapy, especially in idiopathic pulmonary fibrosis (IPF). Current diagnostic methods are imperfect. High-resolution computed tomography has limited resolution, and surgical lung biopsy (SLB) carries risks of morbidity and mortality. Endobronchial optical coherence tomography (EB-OCT) is a low-risk, bronchoscope-compatible modality that images large lung volumes in vivo with microscopic resolution, including subpleural lung, and has the potential to improve the diagnostic accuracy of bronchoscopy for ILD diagnosis. Objectives We performed a prospective diagnostic accuracy study of EB-OCT in patients with ILD with a low-confidence diagnosis undergoing SLB. The primary endpoints were EB-OCT sensitivity/specificity for diagnosis of the histopathologic pattern of usual interstitial pneumonia (UIP) and clinical IPF. The secondary endpoint was agreement between EB-OCT and SLB for diagnosis of the ILD fibrosis pattern. Methods EB-OCT was performed immediately before SLB. The resulting EB-OCT images and histopathology were interpreted by blinded, independent pathologists. Clinical diagnosis was obtained from the treating pulmonologists after SLB, blinded to EB-OCT. Measurements and Main Results We enrolled 31 patients, and 4 were excluded because of inconclusive histopathology or lack of EB-OCT data. Twenty-seven patients were included in the analysis (16 men, average age: 65.0 yr): 12 were diagnosed with UIP and 15 with non-UIP ILD. Average FVC and Dl CO were 75.3% (SD, 18.5) and 53.5% (SD, 16.4), respectively. Sensitivity and specificity of EB-OCT was 100% (95% confidence interval, 75.8–100.0%) and 100% (79.6–100%), respectively, for both histopathologic UIP and clinical diagnosis of IPF. There was high agreement between EB-OCT and histopathology for diagnosis of ILD fibrosis pattern (weighted κ: 0.87 [0.72–1.0]). Conclusions EB-OCT is a safe, accurate method for microscopic ILD diagnosis, as a complement to high-resolution computed tomography and an alternative to SLB.

Abstract: Coronavirus disease 2019 (COVID-19) first emerged in a group of patients who presented with severe pneumonia in Wuhan, China, in December 2019. A novel virus, now called SARSCoV- 2 (Severe Acute Respiratory Syndrome Coronavirus-2), was isolated from lower respiratory tract samples. The current outbreak of infection has spread to over 100 countries and killed more than 340,000 people as of 25th May, 2020. The predominant clinical manifestation of COVID-19 is a respiratory disease- ranging from mild flu-like symptoms to fulminant pneumonia and Acute Respiratory Distress Syndrome (ARDS). Patients with pre-existing cardiovascular risk factors are considered more susceptible to the virus, and these conditions are often worsened by the infection. Furthermore, COVID-19 infection has led to de novo cardiac complications, like acute myocardial injury and arrhythmias. In this review, we have focused on the cardiovascular manifestations of COVID-19 infection that have been reported in the literature so far. We have also outlined the effect of pre-existing cardiovascular disease as well as risk factors on the clinical course and outcomes of COVID-19 infection.