Machine Learning based Early Detection and Monitoring of Parkinson Disease

An extended period of time is necessary to develop Parkinson's disease, a neurodegenerative disorder. The PD can also be identified using human motion analysts, who analyze and identify body movement. According to modern research, changes in speech and vocal sounds are early signs of Parkinson's syndrome. This paper recommends an early detection mechanism created on the examination of speech signals for vocal Parkinson's disease. Treatments are available to ease the symptoms of PD, but the disease cannot be cured. Illness and pain relief treatments available. In mandate to diminish the threat of decease, we are discovering this disease as early as possible. The stenographic and voice recording are both being used to detect PD. IoT cloud is used to diagnose the detected information. The optimal model was developed after combining various algorithms and methods of feature selection. Synthetic Minority Oversampling Technique (SMOTE) can be used to address the extremely unbalanced data-set.