Abstract
We analyse keystroke hold times from typing logs to detect early signs of Parkinson’s disease. We develop a feature that captures the dynamic variation between consecutive keystrokes and demonstrate that it can be be used in a univariate model to perform classification with \(\text {AUC}=0.85\) from only a few hundred keystrokes. This is a substantial improvement on the current baseline. We argue that previously proposed methods are based on overcomplicated models—our simpler method is not only more elegant and transparent but also more effective.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
As given, this will yield a number for each keystroke; it is not explained in Ref. [4] how this measure is then aggregated over the window. Moreover, we note that, contrary to the principles promoted by Giancardo et al., this measure appears to use more than purely hold time data.
- 2.
In fact, each subject in the early PD dataset produced two typing sessions. While training or testing, each typing session is handled independently. If a subject has produced multiple typing sessions then the average nQi is computed to produce a single score.
- 3.
This classification performance is very similar to that obtained using using both \(\langle h \rangle \) and \(\sigma (h)\) as features, whilst the performance using just \(\langle h \rangle \) as a feature is substantially lower.
- 4.
We note again that, unlike the Stdev method, nQi actually appears to use information about the flight time in addition to purely hold time data.
References
Elbaz, A., Carcaillon, L., Kab, S., Moisan, F.: Epidemiology of Parkinson’s disease. Rev. Neurol. 172(1), 14–26 (2016)
Martínez-Martín, P., Gil-Nagel, A., Gracia, L., Gómez, J., Martínez-Sarriés, J., Bermejo, F.: Unified Parkinson’s disease rating scale characteristics and structure. Mov. Disord. 9(1), 76–83 (1994)
Pagan, F.L.: Improving outcomes through early diagnosis of Parkinson’s disease. Am. J. Manag. Care 18(7 Suppl), S176–82 (2012)
Giancardo, L., Sánchez-Ferro, A., Arroyo-Gallego, T., Butterworth, I., Mendoza, C.S., Montero, P., Matarazzo, M., Obeso, J.A., Gray, M.L., Estépar, R.: Computer keyboard interaction as an indicator of early Parkinson’s disease. Sci. Rep. 6, 34468 (2016)
Adams, W.R.: High-accuracy detection of early Parkinson’s Disease using multiple characteristics of finger movement while typing. PLoS One 12(11), e0188226 (2017)
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., Duchesnay, E.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)
Arroyo-Gallego, T., Ledesma-Carbayo, M., Sánchez-Ferro, A., Butterworth, I., Sanchez-Mendoza, C., Matarazzo, M., Montero- Escribano, P., Lopez-Blanco, R., Puertas-Martín, V., Trincado, R., Giancardo, L.: Detection of motor impairment in Parkinson’s disease via mobile touchscreen typing. IEEE Trans. Biomed. Eng. 64(9), 1994–2002 (2017)
Youden, W.J.: Index for rating diagnostic tests. Cancer 3(1), 32–35 (1950)
Christ, M., Kempa-Liehr, A. W., Feindt, M.: Distributed and parallel time series feature extraction for industrial big data applications. arXiv:1610.07717 (2016)
tsfresh, https://github.com/blue-yonder/tsfresh. Accessed 12 Feb 2018
Kaggle: raw data used to predict the onset of Parkinson’s from typing tendencies, https://www.kaggle.com/valkling/tappy-keystroke-data-with-parkinsons-patients. Accessed 1 Aug 2018
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Milne, A., Farrahi, K., Nicolaou, M.A. (2018). Less is More: Univariate Modelling to Detect Early Parkinson’s Disease from Keystroke Dynamics. In: Soldatova, L., Vanschoren, J., Papadopoulos, G., Ceci, M. (eds) Discovery Science. DS 2018. Lecture Notes in Computer Science(), vol 11198. Springer, Cham. https://doi.org/10.1007/978-3-030-01771-2_28
Download citation
DOI: https://doi.org/10.1007/978-3-030-01771-2_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-01770-5
Online ISBN: 978-3-030-01771-2
eBook Packages: Computer ScienceComputer Science (R0)