Author(s): Taha Khan
This paper presents a detailed systematic literatur e review on Parkinson’s disease severity assessment methods bas ed on speech impairment. Techniques that cope up the challenges of voice signal processing in real-time have been reviewed. An analysis on finding relevant acoustic parameters that are robus tly influenced by Parkinson’s disease symptoms and medication-rela ted changes was done. Time-series analysis of voice acoustics, feature extraction and pattern recognition methods are eval uated on the basis of portability and compatibility to mobile de vices. A two- tier review methodology is developed based on the a pplicability of pathological voice recognition methods in real-t ime. The goal is to design an algorithm based on the review synth esis that can tackle real-time constraints in pathological voice recognition for the assessment of Parkinson’s disease severity. Pau se detection, peak to average power rate clipping and zero thresh olding rate calculations are proposed for human voice segmentat ion. Previous work depicts that these methods produce rich voice features in real-time. We suggest that these features may be fu rther processed using wavelet transforms and used with a neural net work for detection and quantification of speech anomalies re lated to Parkinson’s disease.