https://doi.org/10.1140/epjs/s11734-025-01722-9
Regular Article
Leveraging wavelet scattering transform on accelerometry data for classification of Parkinson’s tremor
1
Facultad de Ciencias, Universidad Autónoma de San Luis Potosí, Av. Parque Chapultepec 1570, 78295, San Luis Potosí, San Luis Potosí, Mexico
2
Facultad de Medicina, Universidad Autónoma de San Luis Potosí, Av. Venustiano Carranza 2405, 78210, San Luis Potosí, San Luis Potosí, Mexico
3
Departamento de Bioinformática y Análisis Estadísticos, Instituto Nacional de Perinatología “Isidro Espinosa de los Reyes”, Montes Urales 800, 11000, Ciudad de México, Ciudad de México, Mexico
Received:
31
March
2025
Accepted:
28
May
2025
Published online:
16
June
2025
A novel approach is proposed that leverages accelerometer signals and the wavelet scattering transform (WST) to differentiate between Parkinson’s disease (PD) patients and healthy control participants. WST generates discriminative features by cascading wavelets with nonlinear modulus and averaging operators. Here, scattering coefficients were derived from magnitude signals obtained from triaxial accelerometers placed on the index fingers. Principal component analysis (PCA) was employed to reduce the dimensionality of these coefficients. The processed coefficients were subsequently used as input for training support vector machine (SVM) classifiers. An SVM model was trained for each finger sensor and both models were ensemble for classification. The results demonstrated optimal performance, achieving a maximum value of area under the curve of 0.968 (0.020), with a sensitivity of 99.192(0.990)% and a specificity of 94.367 (3.194)%, respectively, through five-fold cross-validation.The proposed methodology, ensemble WST + PCA + SVM, effectively detects PD using accelerometer signal information, showcasing its potential to advance PD diagnosis and treatment monitoring.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.