The interpretation of pulmonary function tests (PFTs) to diagnose respiratory diseases is built on expert opinion that relies on the recognition of patterns and the clinical context for detection of specific diseases. In this study, we aimed to explore the accuracy and interrater variability of pulmonologists when interpreting PFTs compared with artificial intelligence (AI)-based software that was developed and validated in more than 1500 historical patient cases.120 pulmonologists from 16 European hospitals evaluated 50 cases with PFT and clinical information, resulting in 6000 independent interpretations. The AI software examined the same data. American Thoracic Society/European Respiratory Society guidelines were used as the gold standard for PFT pattern interpretation. The gold standard for diagnosis was derived from clinical history, PFT and all additional tests.The pattern recognition of PFTs by pulmonologists (senior 73%, junior 27%) matched the guidelines in 74.4±5.9% of the cases (range 56–88%). The interrater variability of κ=0.67 pointed to a common agreement. Pulmonologists made correct diagnoses in 44.6±8.7% of the cases (range 24–62%) with a large interrater variability (κ=0.35). The AI-based software perfectly matched the PFT pattern interpretations (100%) and assigned a correct diagnosis in 82% of all cases (p<0.0001 for both measures).The interpretation of PFTs by pulmonologists leads to marked variations and errors. AI-based software provides more accurate interpretations and may serve as a powerful decision support tool to improve clinical practice.
Overall, the application of artificial intelligence has produced encouraging results in the diagnosis of obstructive lung diseases. However, large-scale studies are still required to validate current findings and to boost its adoption by the medical community.
The past 5 years have seen an explosion of interest in the use of artificial intelligence (AI) and machine learning techniques in medicine. This has been driven by the development of deep neural networks (DNNs)—complex networks residing in silico but loosely modelled on the human brain—that can process complex input data such as a chest radiograph image and output a classification such as ‘normal’ or ‘abnormal’. DNNs are ‘trained’ using large banks of images or other input data that have been assigned the correct labels. DNNs have shown the potential to equal or even surpass the accuracy of human experts in pattern recognition tasks such as interpreting medical images or biosignals. Within respiratory medicine, the main applications of AI and machine learning thus far have been the interpretation of thoracic imaging, lung pathology slides and physiological data such as pulmonary function tests. This article surveys progress in this area over the past 5 years, as well as highlighting the current limitations of AI and machine learning and the potential for future developments.
Spirometry is the current gold standard for diagnosing and monitoring the progression of Chronic Obstructive Pulmonary Disease (COPD). However, many current and former smokers who do not meet established spirometric criteria for the diagnosis of this disease have symptoms and clinical courses similar to those with diagnosed COPD. Large longitudinal observational studies following individuals at risk of developing COPD offer us additional insight into spirometric patterns of disease development and progression. Analysis of forced expiratory maneuver changes over time may allow us to better understand early changes predictive of progressive disease. This review discusses the theoretical ability of spirometry to capture fine pathophysiologic changes in early airway disease, highlights the shortcomings of current diagnostic criteria, and reviews existing evidence for spirometric measures which may be used to better detect early airflow impairment.
Background Severe hyperinflation causes detrimental effects such as dyspnea and reduced exercise capacity and is an independent predictor of mortality in COPD patients. Static lung volumes are required to diagnose severe hyperinflation, which are not always accessible in primary care. Several studies have shown that the area under the forced expiratory flow-volume loop (AreaFE) is highly sensitive to bronchodilator response and is correlated with residual volume/total lung capacity (RV/TLC), a common index of air trapping. In this study, we investigate the role of AreaFE% (AreaFE expressed as a percentage of reference value) and conventional spirometry parameters in indicating severe hyperinflation. Materials and methods We used a cohort of 215 individuals with COPD. The presence of severe hyperinflation was defined as elevated air trapping (RV/TLC >60%) or reduced inspiratory fraction (inspiratory capacity [IC]/TLC <25%) measured using body plethysmography. AreaFE% was calculated by integrating the maximal expiratory flow-volume loop with the trapezoidal rule and expressing it as a percentage of the reference value estimated using predicted values of FVC, peak expiratory flow and forced expiratory flow at 25%, 50% and 75% of FVC. Receiver operating characteristics (ROC) curve analysis was used to identify cut-offs that were used to indicate severe hyperinflation, which were then validated in a separate group of 104 COPD subjects. Results ROC analysis identified cut-offs of 15% and 20% for AreaFE% in indicating RV/TLC >60% and IC/TLC <25%, respectively (N=215). On validation (N=104), these cut-offs consistently registered the highest accuracy (80% each), sensitivity (68% and 75%) and specificity (83% and 80%) among conventional parameters in both criteria of severe hyperinflation. Conclusion AreaFE% consistently provides a superior estimation of severe hyperinflation using different indices, and may provide a convenient way to refer COPD patients for body plethysmography to address static lung volumes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.