Presenter of 1 Presentation
O051 - EVALUATION OF MACHINE LEARNING TO DETECT ADVENTITIOUS LUNG SOUNDS USING DIGITAL AUSCULTATION TO AID CHILDHOOD PNEUMONIA DIAGNOSIS (ID 280)
Abstract
Background
IMCI guidelines for childhood pneumonia diagnosis have high sensitivity but low specificity. A digital stethoscope with an automated machine-learning algorithm for classifying lung sounds may improve diagnostic performance. We aimed to evaluate agreement between digital stethoscope recorded lung sound classifications generated from an automated machine learning algorithm with a paediatrician listening panel from children receiving care at community clinics in rural Bangladesh.
Methods
Government community health workers recorded lung sounds from four chest positions using a novel digital stethoscope in children under-5 with cough and/or difficult breathing at first-level community clinics in Bangladesh from November 2019 to December 2020. A trained paediatrician listening panel classified recorded lung sounds into normal, crackles, wheeze, crackles and wheeze, or uninterpretable. A machine learning algorithm classified recorded sounds into the same categories, which were compared with panel classifications.
Results
Of 2434 children screened, 990 were enrolled. Compared to paediatricians, the sensitivity, specificity, and positive and negative predictive values of detecting abnormal sounds (wheeze and/or crackles) by the machine learning algorithm were 61.8 (95%CI: 55.7, 67.6), 60.7 (56.6, 64.6), 41.8 (36.9, 46.8), and 77.6 (73.6, 81.3) among all enrolled children, and 63.5 (54.5, 71.9), 66.2 (60.1, 73.1), 52.7 (44.4, 60.8), and 75.9 (69.2, 81.8) among children with IMCI defined pneumonia.
Conclusions
This study shows an automated algorithm had moderate sensitivity and specificity for classifying lung sounds as either abnormal or normal when using a paediatric listening panel as the reference. Agreement between the machine learning algorithm and paediatric listening panel modestly increased among children with IMCI-defined pneumonia.