Predicting Parkinson ’ s Disease Severity from Patient Voice Features

@inproceedings{Genain2014PredictingP,
  title={Predicting Parkinson ’ s Disease Severity from Patient Voice Features},
  author={Nicolas Genain and Madeline Huberth and Roshan Vidyashankar},
  year={2014}
}
This paper describes the machine learning methods and modeling used to predict continuous measures of Parkinson’s Disease Severity from voice recordings of patients. Two datasets are analyzed. Bagged decision trees (random forests) resulted in an improvement on the previous model accuracy for one dataset, predicting severity measures at 2% accuracy on a 0-176 scale. Other methods are described for both datasets, as well as the limitations of each. 

Similar Papers

References

Publications referenced by this paper.
SHOWING 1-10 OF 10 REFERENCES

High accuracy discrimination of Parkinson's disease participants from healthy controls using smartphones

  • 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2014
VIEW 1 EXCERPT

Early detection of Parkinson’s disease by voice analysis.

of Haifa
  • ScienceDaily. ScienceDaily,
  • 2010
VIEW 1 EXCERPT

Random forests

M. A. Little, P. E. McSharry, E. J. Hunter, J. Spielman, L. O. Ramig
  • Machine Learning
  • 2001

Comparing partitions in Journal of Classification pp

Hubert, Lawrence, Arabie, Phipps
  • 1985
VIEW 1 EXCERPT