Skip to Main content Skip to Navigation
Journal articles

Data Imputation and Compression For Parkinson's Disease Clinical Questionnaires

Abstract : Medical questionnaires are a valuable source of information but are often difficult to analyse due to both their size and the high possibility of having missing values. This is a problematic issue in biomedical data science as it may complicate how individual questionnaire data is represented for statistical or machine learning analysis. In this paper, we propose a deeply-learnt residual autoencoder to simultaneously perform non-linear data imputation and dimensionality reduction. We present an extensive analysis of the dynamics of the performances of this autoencoder regarding the compression rate and the proportion of missing values. This method is evaluated on motor and non-motor clinical questionnaires of the Parkinson's Progression Markers Initiative (PPMI) database and consistently outperforms linear coupled imputation and reduction approaches.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02570967
Contributor : Maxime Peralta <>
Submitted on : Sunday, April 11, 2021 - 1:27:17 PM
Last modification on : Monday, May 3, 2021 - 3:53:48 PM

File

PatiNAE_hal.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution - NonCommercial - ShareAlike 4.0 International License

Identifiers

Collections

Citation

Maxime Peralta, Pierre Jannin, Claire Haegelen, John Baxter. Data Imputation and Compression For Parkinson's Disease Clinical Questionnaires. Artificial Intelligence in Medicine, Elsevier, 2021, 114, pp.102051. ⟨10.1016/j.artmed.2021.102051⟩. ⟨hal-02570967v3⟩

Share

Metrics

Record views

101

Files downloads

74