AusculSensor: Um sistema para auxílio na Ausculta Pulmonar

Abstract

Respiratory diseases are one of the main causes of death in the world, being among the global priorities proposed by the World Health Organization (WHO). Thus, respiratory physiotherapy is one of the main treatments for lung diseases. Pulmonary auscultation is extremely important, as it is the main method of physical examination of the chest that allows a better analysis of lung functioning for the detection of respiratory diseases. This procedure is used due to its great functionality, as it is a non-invasive procedure and also has a rapid detection of abnormalities. Due to the Coronavirus Disease (COVID-19), this technique has become essential to determine the severity of the disease, as it is possible to identify respiratory changes such as snoring and crackles, as well as other signs of severity in acute respiratory conditions. Therefore, this study developed a system capable of assisting in the application of the pulmonary auscultation technique, through a device capable of capturing sound data during the auscultation stage. This system relies on the integration of hardware and software, using a microphone sensor node to capture data from respiratory sounds coupled to a stethoscope and a microcontroller board to process the data from the sensor. In the analysis screen, it is was possible to verify the data of the sounds obtained with this system through an amplitude versus time graph. The modeling of this software started from the definition of requirements, screen prototyping, the survey of use cases, sequence and class diagrams, and database. In addition, the system allowed students and health professionals to carry out training to aid in the acoustic identification of lung sounds. This learning step was carried out through the pulmonary sounds auscultation training module and the implemented pulmonary quiz. In addition, experiments were carried out with physical therapists, evaluating all the functions developed in the solution. The results obtained showed a high potential of applicability for the proposed problem.