Parkinson’s Disease (PD) is a degeneration of the brain leading to cognitive problems. It’s one of the most frequent neurodegenerative diseases and it affects more than 10 million globally. It causes trembling, stiffness, and issues with walking, balancing, and coordinating. Symptoms come gradually with sleep difficulties, depression, exhaustion, and memory problems. Yet, the diagnosis is still difficult to evaluate, especially in its early stages. PD disease manifests itself differently in each individual. In this project, the goal is to demonstrate if a correlation exists between PD diagnosing positively and the patient’s ability to speak distinctly. More info here: www.parkinson.org
It’s a supervised learning problem with labeled data (we know the input and output). Involves building a model to estimate or predict an output based on one or more inputs. We use classification. The most popular algorithms are : Random Forest KNN AdaBoost Decision Tree Naive Bayes
It’s a dataset composed of biomedical voice measurements from 50 people There 6 recordings per patient 195 rows of data and 23 features for the columns. The features are numeric except for the first column “name” Let’s go deeper. Patients number 7, 10, 13, 17, 42, 43, 49, and 50 do not have PD (8 patients), and others do.
In the first column “name” listed “phon_R01_S01_1” stands for the first recording of the first patient “phon_R01_S01_2” stands for the second recording of the first patient etc…
MDVP: Fo(Hz) - Average vocal fundamental frequency MDVP: Fhi(Hz) - Maximum vocal fundamental frequency MDVP: Flo(Hz) - Minimum vocal fundamental frequency MDVP: Jitter(%), MDVP: Jitter(Abs), MDVP: RAP, MDVP: PPQ Jitter:DDP - Several measures of variation in fundamental frequency MDVP:Shimmer,MDVP:Shimmer(dB),Shimmer:APQ3,Shimmer:APQ5,MDVP:APQ,Shimmer:DDA - Several measures of variation in amplitude NHR,HNR -
Health status of the subject 1 = unhealthy (PD) 0 = healthy
RPDE, D2 - Two nonlinear dynamical complexity measures DFA - Signal fractal scaling exponent spread1,spread2,PPE - Three nonlinear measures of fundamental frequency variation