Artificial intelligence in pulmonary medicine: computer vision, predictive model and COVID-19.

Artificial intelligence (AI) is transforming healthcare delivery. The digital revolution in medicine and healthcare information is prompting a staggering growth of data intertwined with elements from many digital sources such as genomics, medical imaging and electronic health records. Such massive growth has sparked the development of an increasing number of AI-based applications that can be deployed in clinical practice. Pulmonary specialists who are familiar with the principles of AI and its applications will be empowered and prepared to seize future practice and research opportunities. The goal of this review is to provide pulmonary specialists and other readers with information pertinent to the use of AI in pulmonary medicine. First, we describe the concept of AI and some of the requisites of machine learning and deep learning. Next, we review some of the literature relevant to the use of computer vision in medical imaging, predictive modelling with machine learning, and the use of AI for battling the novel severe acute respiratory syndrome-coronavirus-2 pandemic. We close our review with a discussion of limitations and challenges pertaining to the further incorporation of AI into clinical pulmonary practice.

Driving pressure is not associated with mortality in mechanically ventilated patients without ARDS.

BACKGROUND: In patients with acute respiratory distress syndrome (ARDS), low tidal volume ventilation has been associated with reduced mortality. Driving pressure (tidal volume normalized to respiratory system compliance) may be an even stronger predictor of ARDS survival than tidal volume. We sought to study whether these associations hold true in acute respiratory failure patients without ARDS. METHODS: This is a retrospectively cohort analysis of mechanically ventilated adult patients admitted to ICUs from 12 hospitals over 2 years. We used natural language processing of chest radiograph reports and data from the electronic medical record to identify patients who had ARDS. We used multivariable logistic regression and generalized linear models to estimate associations between tidal volume, driving pressure, and respiratory system compliance with adjusted 30-day mortality using covariates of Acute Physiology Score (APS), Charlson Comorbidity Index (CCI), age, and PaO2/FiO2 ratio. RESULTS: We studied 2641 patients; 48% had ARDS (n = 1273). Patients with ARDS had higher mean APS (25 vs. 23, p < .001) but similar CCI (4 vs. 3, p = 0.6) scores. For non-ARDS patients, tidal volume was associated with increased adjusted mortality (OR 1.18 per 1 mL/kg PBW increase in tidal volume, CI 1.04 to 1.35, p = 0.010). We observed no association between driving pressure or respiratory compliance and mortality in patients without ARDS. In ARDS patients, both ΔP (OR1.1, CI 1.06-1.14, p < 0.001) and tidal volume (OR 1.17, CI 1.04-1.31, p = 0.007) were associated with mortality. CONCLUSIONS: In a large retrospective analysis of critically ill non-ARDS patients receiving mechanical ventilation, we found that tidal volume was associated with 30-day mortality, while driving pressure was not.

Pleural Effusions at First ED Encounter Predict Worse Clinical Outcomes in Patients With Pneumonia.

BACKGROUND: Pleural effusions are present in 15% to 44% of hospitalized patients with pneumonia. It is unknown whether effusions at first presentation to the ED influence outcomes or should be managed differently. METHODS: We studied patients in seven hospital EDs with International Statistical Classification of Disease and Health Related Problems-Version 9 codes for pneumonia, or empyema, sepsis, or respiratory failure with secondary pneumonia. Patients with no confirmatory findings on chest imaging were excluded. Pleural effusions were identified with the use of radiographic imaging. RESULTS: Over 24 months, 4,771 of 458,837 adult ED patients fulfilled entry criteria. Among the 690 (14.5%) patients with pleural effusions, their median age was 68 years, and 46% were male. Patients with higher Elixhauser comorbidity scores (OR, 1.13 [95% CI, 1.09-1.18]; P < .001), brain natriuretic peptide levels (OR, 1.20 [95% CI, 1.12-1.28]; P < .001), bilirubin levels (OR, 1.07 [95% CI, 1.00-1.15]; P = .04), and age (OR, 1.15 [95% CI, 1.09-1.21]; P < .001) were more likely to have parapneumonic effusions. In patients without effusion, electronic version of CURB-65 (confusion, uremia, respiratory rate, BP, age ≥ 65 years accurately predicted mortality (4.7% predicted vs 5.0% actual). However, eCURB underestimated mortality in those with effusions (predicted 7.0% vs actual 14.0%; P < .001). Patients with effusions were more likely to be admitted (77% vs 57%; P < .001) and had a longer hospital stay (median, 2.8 vs 1.3 days; P < .001). After severity adjustment, the likelihood of 30-day mortality was greater among patients with effusions (OR, 2.6 [CI, 2.0-3.5]; P < .001), and hospital stay was disproportionately longer (coefficient, 0.22 [CI, 0.14-0.29]; P < .001). CONCLUSIONS: Patients with pneumonia and pleural effusions at ED presentation in this study were more likely to die, be admitted, and had longer hospital stays. Why parapneumonic effusions are associated with adverse outcomes, and whether different management of these patients might improve outcome, needs urgent investigation.

Pneumococcal urinary antigen test use in diagnosis and treatment of pneumonia in seven Utah hospitals.

The pneumocococcal urine antigen test increases specific microbiological diagnosis over conventional culture methods in pneumonia patients. Data are limited regarding its yield and effect on antibiotic prescribing among patients with community-onset pneumonia in clinical practice. We performed a secondary analysis of 2837 emergency department patients admitted to seven Utah hospitals over 2 years with international diagnostic codes version 9 codes and radiographic evidence of pneumonia. Mean age was 64.2 years, 47.2% were male and all-cause 30-day mortality was 9.6%. Urinary antigen testing was performed in 1110 (39%) patients yielding 134 (12%) positives. Intensive care unit patients were more likely to undergo testing, and have a positive result (15% versus 8.8% for ward patients; p<0.01). Patients with risk factors for healthcare-associated pneumonia had fewer urinary antigen tests performed, but 8.4% were positive. Physicians changed to targeted antibiotic therapy in 20 (15%) patients, de-escalated antibiotic therapy in 76 patients (57%). In 38 (28%) patients, antibiotics were not changed. Only one patient changed to targeted therapy suffered clinical relapse. Length of stay and mortality were lower in patients receiving targeted therapy. Pneumococcal urinary antigen testing is an inexpensive, noninvasive test that favourably influenced antibiotic prescribing in a "real world", multi-hospital observational study.