Novel Artificial Intelligence-Based Technology to Diagnose Asthma Using Methacholine Challenge Tests
Noeul Kang, Kyung Hyun Lee, Sangwon Byun, Jin-Young Lee, Dong‐Chull Choi, Byung‐Jae Lee
Abstract
PURPOSE: The methacholine challenge test (MCT) has high sensitivity but relatively low specificity for asthma diagnosis. This study aimed to develop and validate machine learning (ML) models to improve the diagnostic performance of MCT for asthma. METHODS: ) values obtained during MCT. RESULTS: were included as inputs. CONCLUSIONS: ≤ 16 mg/mL. The novel technology could be used to enhance the clinical diagnosis of asthma.
Topics & Concepts
AsthmaMedicineReceiver operating characteristicConfidence intervalVital capacityRandom forestLogistic regressionArtificial neural networkMachine learningArtificial intelligenceStatisticsInternal medicineMathematicsComputer scienceLung functionLungDiffusing capacityAsthma and respiratory diseasesRespiratory and Cough-Related ResearchInhalation and Respiratory Drug Delivery