Lung sounds classification with artificial intelligence

Introduction: The characteristics of lung sounds and their diagnoses are an integral part of lung pathology. However, its biggest drawback is its subjectivity. The results depend on the experience and ability of the doctor to perceive and distinguish pathologies in the sounds heard via a stethoscope. Aim: To Develop a new method of automatic sound analysis based on convolutional neural networks (CNN) and implementing it in a system that uses an electronic stethoscope to capture respiratory sounds and assign them the correct auscultation. Materials and methods: The auscultation recording was collected from 57 patients admitted to our departement, performed by the same pulmonologist using Littmann 3200 electronic stethoscopes, 4 auscultations foci for each patient. Each recording had a description and two independent verifications by two different pulmonologists. The results were compared with an artificial intelligence model developed by a Tunisian startup in the classification of Lung sounds. Results: The pulmonologists had the same classification of lung sound in 96%. For the rest 4% the recording was analyzed by an acoustician experienced in signal recognition. These records were qualified GOLDEN STANDARD. After injection of the collected sound in the IA model the F1-score= 0.90. Conclusion: The Artificial intelligence of the lung sound classification will be a useful means to facilitate clinical decision making.