Avaliação de sinais acústicos para o pré-diagnóstico de covid-19

• Autor(a) principal: Nunes, Gabriel Andrey Perego
• Data de Publicação: 2022
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/216137
• topic: Signal processing; Wavelet transform; Pathology identification; Teager operator; Processamento de sinais - Técnicas digitais; Transformada de Wavelets; Identificação de patologia; Operador de Teager

The signals produced by the human body have been the subject of researches on indicators related to different diseases. With the outbreak of the Sars-Cov-2 pandemic in 2020, there is a unique opportunity for signal processing to contribute to the identification of the pathology. The first papers on the new virus were based on recent studies of other diseases that have similar symptoms to those caused by the coronavirus. Since the disease has common symptoms in patients and these symptoms can cause some degree of impairment in the vocal system and generate irregular vibrations, this study aims to identify Covid-19 signatures in audios with sustained vowels through analysis of the energy in the sound waves. By using Wavelet transform and Teager operator, it was possible to identify, with statistical methods, unique characteristics of healthy and infected individuals, which can be used in the creation of classifiers to identify the pathology. It was found that infected people seems to have a vector with an average value of 0.00034 while healthy people produce a value of 0.01665. Also, it was possible to analyze the difference between the minimum and maximum values found in the vectors.