Assessment of Handwriting in Patients With Parkinson's Disease Using Non-Intrusive Tasks
Gallo-Aristizabal JD, Escobar-Grisales D, Rios-Urrego CD, Perez Toro PA, Nöth E, Maier A, Orozco Arroyave JR (2023)
Publication Type: Conference contribution
Publication year: 2023
Journal
Publisher: IEEE Computer Society
Book Volume: 2023-April
Conference Proceedings Title: Proceedings - International Symposium on Biomedical Imaging
Event location: Cartagena, COL
ISBN: 9781665473583
DOI: 10.1109/ISBI53787.2023.10230617
Abstract
This study presents two approaches for modeling the handwriting of Parkinson's Disease (PD) patients and Healthy Control (HC) subjects. One approach is based on digit-embeddings generated from a CNN architecture pre-trained with information from the MNIST corpus. The second approach consists of the computation of statistical functionals of dynamics signal collected with the digital tablet, namely azimuth, pressure, altitude, and vertical distance. The experiments are based on writing the ten digits (from 0 to 9), which is a task commonly performed in daily life activities, making this approach closer to a non-intrusive evaluation. According to the results, the accuracy of the classification between PD patients vs. HC improved from 71.8% to 74.5% when information from images is combined with the functionals of the vertical distance signal.
Authors with CRIS profile
Paula Andrea Perez Toro
Technische Fakultät
Elmar Nöth
Professur für Informatik (Mustererkennung)
Andreas Maier
Lehrstuhl für Informatik 5 (Mustererkennung)
Juan Rafael Orozco Arroyave
Professur für Informatik (Mustererkennung)
Additional Organisation(s)
Involved external institutions
Colombia (CO)How to cite
APA:
Gallo-Aristizabal, J.D., Escobar-Grisales, D., Rios-Urrego, C.D., Perez Toro, P.A., Nöth, E., Maier, A., & Orozco Arroyave, J.R. (2023). Assessment of Handwriting in Patients With Parkinson's Disease Using Non-Intrusive Tasks. In Proceedings - International Symposium on Biomedical Imaging. Cartagena, COL: IEEE Computer Society.
MLA:
Gallo-Aristizabal, J. D., et al. "Assessment of Handwriting in Patients With Parkinson's Disease Using Non-Intrusive Tasks." Proceedings of the 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023, Cartagena, COL IEEE Computer Society, 2023.
BibTeX: Download