METHODS AND APPARATUSES FOR DIAGNOSING PARKINSONS DISEASE USING SPEECH DATA BASED ON CLASS PROBABILITY OUTPUT NETWORK

摘要

The present invention relates to a method and apparatus for diagnosing Parkinson''s disease using speech data based on a class probability output network, and the method and apparatus for diagnosing Parkinson''s disease using speech data based on a class probability output network according to an embodiment of the present invention include, Normalizing the voice features extracted from , linearly combining the normalized voice features and normalizing the linearly combined output, determining the beta parameter of the normalized output distribution for disease and normal data to measure uncertainty calculating a (Uncertainty Measure) value, estimating a significance probability (p-value) for examining a disease class and a normal class using the determined beta parameter, and the estimated value according to the calculated uncertainty measure value and diagnosing the disease of the diagnosis target using the significance probability.