Artificial Intelligent (AI) Clinical Edge for Voice disorder Detection

Computerized detection of voice disorders have attracted considerable academic and clinical interest in the hope of providing an effective screening method for voice diseases before endoscopic confirmation. The goal of this paper is to apply neural networks and Machin learning techniques to detect pathological voice and classify three disordered categories from acoustic waveforms collected on the mobile phone. From a health science perspective, a pathological status of the human voice can substantially reduce quality of life and occupational performance, which results in considerable costs for both the patient and society. The paper summarizes the various techniques and feature engineering processes that we have applied for the Voice Data collected for classification of voice disorders. We have used Mel scaled spectrograms and MFCC components as audio features to train various Neural Network Architectures. We have trained a 5layer plain network, 5-layer CNN and RNN. We discuss the challenges faced and solutions to improve model performance, model parameter tuning and model evaluation.