Application of logistic regression algorithm in the diagnosis of expression disorder in Parkinson's disease

Parkinson's disease (PD) is a common neurodegenerative disease with low mortality but high disability rate, and the prevalence gradually increases with age. “Hypomimia” is considered to be one of the common symptoms of Parkinson's patients. The patient seldom blinks, his eyes turn less, and his expression is dull. Even if he makes an expression intentionally, he appears very stiff, as if he is wearing a mask. Computer image processing and machine learning help patients discover their disease and obtain treatment in a relatively short period of time. For this reason, based on the Facial Action Coding System (FACS), this paper proposes a diagnosis and evaluation method of Parkinson's disease expression disorder based on facial action units. A quantitative evaluation method of facial expression behavior based on facial action units is proposed. A differential diagnosis model for Parkinson's disease is established. The validity of recognition model is verified.The severity of Hypomimia symptoms is graded. The relationship between facial expression disorder and disease grade in patients is explored. The accuracy of the optimal threshold value of the logistic regression classifier is 90.06%.The results have practical significance and application value for objective diagnosis of Parkinson's disease.