On the use of cepstral coefficients and Multilayer Perceptron Networks for vocal fold edema diagnosis

Laryngeal diseases affect many professionals who use their voices as the main working tool, such as teachers, singers, radio and TV presenters, among others. Advanced diagnosis techniques of these diseases are typically invasive, causing much discomfort to the patient. In recent years techniques of digital voice processing have been investigated to obtain non-invasive systems to aid the diagnosis by a specialist. This work proposes a method of analysis that employs cepstral coefficients to represent the voice signals, and multilayer perceptron neural networks for discrimination among normal voice, voices affected by local fold Edema and voices affected by other pathologies (nodules, cysts and paralysis). An experimental evaluation of the method has demonstrated that this is a promising approach to the problem, reaching a correct classification rate above 99% for normal voice, 96% for Edema and 93% for other pathologies.