Unsupervised morphological segmentation for detecting Parkinson?s disease

IBÁÑEZ, AGUSTÍN; OROZCO-ARROYAVE, JUAN RAFAEL; PLOSECKI, PABLO; CECCHI, GUILLERMO; SKODDA, SABINA; GARCÍA, ADOLFO M.; HESSE, EUGENIA; ROGG, KATHARINA; EYIGOZ, ELIF

Nueva Orleans

Workshop; AAAI-18 Workshop on Artificial Intelligence Applied to Assistive Technologies and Smart Environments; 2018

AAAI

Computer-aided detection of aging-related neuro-degenerative disorders, such as Parkinson?s disease, has gained a lot of interest as the population ages. We propose a novel method of feature extraction utilizing unsupervised morphological segmentation for accessing the complexity of a speaker?s language. According to our experiments on German, our method can classify patients vs. healthy controls with 81 percent accuracy, and estimate the neurological state of PD patients with Pearson correlation of 0.44 with respect to the unified Parkinson?s disease rating scale. Our work is the first study to show that unsupervised morphological segmentation can be used for automatic detection of a neurological disorder.