lsvt

OpenML1484MaRDI QIDQ6034062

OpenML dataset with id 1484

No author found.

Full work available at URL: https://api.openml.org/data/v1/download/1590985/lsvt.arff

Upload date: 22 May 2015

Dataset Characteristics

Number of classes: 2
Number of features: 311 (numeric: 310, symbolic: 1 and in total binary: 1 )
Number of instances: 126
Number of instances with missing values: 0
Number of missing values: 0

Description

Author: Athanasios Tsanas

Source: UCI

Please cite: A. Tsanas, M.A. Little, C. Fox, L.O. Ramig: Objective automatic assessment of rehabilitative speech treatment in Parkinsons disease, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 22, pp. 181-190, January 2014

Dataset title

laLSVT Voice Rehabilitation Data Set

Source:

The dataset was created by Athanasios Tsanas (tsanasthanasis '@' gmail.com) of the University of Oxford.

Abstract: 126 samples from 14 participants, 309 features. Aim: assess whether voice rehabilitation treatment lead to phonations considered 'acceptable' or 'unacceptable' (binary class classification problem).

Data Set Information:

The original paper demonstrated that it is possible to correctly replicate the experts' binary assessment with approximately 90% accuracy using both 10-fold cross-validation and leave-one-subject-out validation. We experimented with both random forests and support vector machines, using standard approaches for optimizing the SVM's hyperparameters. It will be interesting if researchers can improve on this finding using advanced machine learning tools.

Details for the dataset can be found on the following paper. A. Tsanas, M.A. Little, C. Fox, L.O. Ramig: “Objective automatic assessment of rehabilitative speech treatment in Parkinson’s disease”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 22, pp. 181-190, January 2014

A freely available preprint is availabe from the first author's website.

Attribute Information:

Each attribute (feature) corresponds to the application of a speech signal processing algorithm which aims to characterise objectively the signal. These algorithms include standard perturbation analysis methods, wavelet-based features, fundamental frequency-based features, and tools used to mine nonlinear time-series. Because of the extensive number of attributes we refer the interested readers to the relevant papers for further details.

Relevant Papers:

The dataset was introduced in: A. Tsanas, M.A. Little, C. Fox, L.O. Ramig: “Objective automatic assessment of rehabilitative speech treatment in Parkinson’s disease”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 22, pp. 181-190, January 2014

Further details about the speech signal processing algorithms can be found in:

A. Tsanas, Accurate telemonitoring of Parkinson’s disease symptom severity using nonlinear speech signal processing and statistical machine learning, D.Phil. (Ph.D.) thesis, University of Oxford, UK, 2012

A. Tsanas, M.A. Little, P.E. McSharry, L.O. Ramig: “Nonlinear speech analysis algorithms mapped to a standard metric achieve clinically useful quantification of average Parkinson’s disease symptom severity”, Journal of the Royal Society Interface, Vol. 8, pp. 842-855, 2011

A. Tsanas, M.A. Little, P.E. McSharry, L.O. Ramig: “New nonlinear markers and insights into speech signal degradation for effective tracking of Parkinson’s disease symptom severity”, International Symposium on Nonlinear Theory and its Applications (NOLTA), pp. 457-460, Krakow, Poland, 5-8 September 2010

Preprints are available on the first author's website.