PDetect-Parkinson’s Disease Detection Using Speech Features

Abstract

Parkinson’s syndrome is a degenerative,progressive neurological condition. The substantia nigra(a part of brain) nerve cells in tiny bundles are mostly affected.When everything is working properly, these cells produce dopamine, a chemical (neurotransmitter) that carries information between brain areas to coordinate smooth and balanced muscular movement. Because these nerve cells die as a result of Parkinson’s disease,body motions are altered. Early Parkinson’s disease intervention is crucial for slowing the disease’s course. By allowing patients to have access to information Condition modifying therapy is a type of medication that is used to treat a disease. To ameliorate the symptoms of Parkinson’s disease, computational algorithms that utilise a set of data containing medical information about the disease. This will aid the number of people who seek to identify a risk early on. Research conducted on the Parkinson’s disease (PD) detection voice impediment has been proven to exist, is associated with symptoms. It is seen in 90% of people with early-stage Parkinson’s syndrome. Therefore, we are interested in applying the vowel function to computer-aided Clinical recognition and remote patient monitoring of patients with parkinson. The prognosis for Parkinson’s disease is poor. However, both genetic and environmental factors may be involved. For this project, I’ve opted to focus on medicine and use vocalisation data to classify whether or not a person has Parkinson’s disease. For background, it is a degenerative brain disorder that causes both movement and cognitive impairment. As a result, it’s realistic to infer that a patient’s capacity to, as these abilities deteriorate, they develop Parkinson’s disease. Compared to older Machine Learning models, the proposed methodology greatly improves the accuracy of PD identification

Keywords: Parkinson disease, Speech processing, Machine learning, Acoustic features