DSFM (version 1.0.1)

Parkinsons_Features: Parkinson's Disease Voice Features Dataset

Description

A dataset containing biomedical voice measurements from people with Parkinson's disease and healthy controls. The goal is to analyze voice signal features for detecting and monitoring Parkinson's disease.

Usage

data(Parkinsons_Features)

Format

A data frame with 5,876 observations on 22 variables. Each row corresponds to a voice recording from a subject.

subject_id	Identifier for the subject (factor or character)
age	Age of the subject (numeric)
sex	Sex of the subject (factor: Male/Female)
test_time	Time of test (numeric, days since baseline)
motor_UPDRS	Unified Parkinson's Disease Rating Scale motor score (numeric)
total_UPDRS	Total UPDRS score (numeric)
Jitter	Measure of frequency variation (numeric)
Shimmer	Measure of amplitude variation (numeric)
NHR	Noise-to-harmonics ratio (numeric)
HNR	Harmonics-to-noise ratio (numeric)
RPDE	Recurrence period density entropy (numeric)
DFA	Detrended fluctuation analysis (numeric)
PPE	Pitch period entropy (numeric)
...	Additional voice signal features and measurements (numeric)

All features are numerical except for identifiers and categorical variables.

Details

This dataset was collected from subjects with Parkinson's disease and healthy controls. Multiple biomedical voice measurements were recorded over time to evaluate disease progression.

The features include various jitter, shimmer, noise, and entropy measures extracted from sustained vowel phonations.

The dataset is widely used for classification and regression models aiming to predict Parkinson's disease severity or presence.

References

Tsanas, A., Little, M.A., McSharry, P.E., & Ramig, L.O. (2010). Accurate telemonitoring of Parkinson's disease progression by noninvasive speech tests. IEEE Transactions on Biomedical Engineering, 57(4), 884–893.

Examples

data(Parkinsons_Features)

if (all(startsWith(names(Parkinsons_Features), "V"))) {
  colnames(Parkinsons_Features) <- Parkinsons_Features[1, ]
  Parkinsons_Features <- Parkinsons_Features[-1, ]
}

Parkinsons_Features[] <- lapply(Parkinsons_Features, type.convert, as.is = TRUE)

summary(Parkinsons_Features$motor_UPDRS)
boxplot(motor_UPDRS ~ sex, data = Parkinsons_Features,
        main = "Motor UPDRS by Sex", ylab = "Motor UPDRS")