Artificial Intelligence Techniques for the Identification of the Severity of Speech Disorder in Patients with Parkinson Tremors

Parkinson Disease (PD) is one of the most frequent neurodegenerative diseases worldwide. In addition to motor disorders, patients normally suffer from a speech disorder named Dysarthria. Usually the degree of difficulty (severity) of this speech disorder is classified by one of the different possible scales like the National technical Institute of the Deaf (NTID) scale. These ratings are currently done by auditive assessments, which are very time consuming and costly and uncomfortable for both patients and assessors. The aim of this research is to come with an objective automated tool to analyse the speech signals for PD patients and to estimate the severity of the speech disorder. Different frequency domain and time domain features are to be obtained and used as inputs to an artificial neural network in the classification scheme. Artificial intelligence methods (especially computer aided diagnosis and artificial neural networks) can be employed. These adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized outputs. In this project, different neural networks from multi-layer Feed-Forward Back-Propagation NN to Deep Learning NN are to be used for identification of level of Dysarthria. This will provide an automatic simple method to replace classical assessment methods. NN will help also in identification based on short length speech segments.

Parkinson Disease (PD) is one of the most frequent neurodegenerative diseases worldwide. In addition to motor disorders, patients normally suffer from a speech disorder named Dysarthria. Usually the degree of difficulty (severity) of this speech disorder is classified by one of the different possible scales like the National technical Institute of the Deaf (NTID) scale. These ratings are currently done by auditive assessments, which are very time consuming and costly and uncomfortable for both patients and assessors. The aim of this research is to come with an objective automated tool to analyse the speech signals for PD patients and to estimate the severity of the speech disorder. Different frequency domain and time domain features are to be obtained and used as inputs to an artificial neural network in the classification scheme. Artificial intelligence methods (especially computer aided diagnosis and artificial neural networks) can be employed. These adaptive learning algorithms can handle diverse types of medical data and integrate them into categorized outputs. In this project, different neural networks from multi-layer Feed-Forward Back-Propagation NN to Deep Learning NN are to be used for identification of level of Dysarthria. This will provide an automatic simple method to replace classical assessment methods. NN will help also in identification based on short length speech segments.