Examination of Mel-frequency Cepstral Coefficients features for Parkinson's Disease telediagnosis

Abstract

The focus for 5G networks is on high fidelity networks with extreme precision, making complete telediagnostic and telehealth realizations a possibility. The characterization of Parkinson’s disease (PD) is based on motor and non-motor functions of the body. Speech issues in PD patients identify with motor control, not to lost the phonetic learning required to make prosodic qualifications. Speech discourse issues are referred to as one of the most punctual markers for PD and clinical conclusion takes up to five years to identify PD. This means s pitch can be used as one of the biomarkers for PD. Attention is given to prosodic, spectral, vocal tract and excitation source speech features. Fundamental frequency, Pitch Period Entropy (PPE), jitter, shimmer, Melfrequency Cepstral Coefficients (MFCC) and Hidden Markov Model (HMM) are the parameters used in this respect