Nicolo' Formento Moletta
IMPROVING EFFICIENCY AND ROBUSTNESS IN INFANT EEG PREPROCESSING An Optimized Pipeline.
Rel. Gabriella Olmo, Ghislaine Dehaene-Lambertz, Lorenzo Jhunlyn, Ana Flo'. Politecnico di Torino, Master of science program in Biomedical Engineering, 2025
|
Preview |
PDF (Tesi_di_laurea)
- Thesis
Licence: Creative Commons Attribution Non-commercial No Derivatives. Download (6MB) | Preview |
Abstract
APICE-Py constitute the fully open-source Python counterpart of the APICE pipeline, a MATLAB-based tool for infant Electroencephalogram (EEG) preprocessing. It is devel??oped to enhance accessibility, foster collaboration among research teams, and integrate seamlessly with modern machine learning tools. The difference between the programming languages, required key modifications, notably the parallelization of the Spherical Spline Interpolation (SSI), which dramatically im??proved computational efficiency for large datasets when compared to the original func??tion of the MNE-Python library. While numerical differences in filter implementations led APICE-Py to be more conservative in the number of retained epochs, the valida??tion demonstrated comparable performance. Using the Standardized Measurement Error (SME), no significant statistical differences were found in the extracted Evoked Response Potential (ERP)s between the Python and MATLAB versions.
Computational time anal??yses confirmed that the parallelized Python version achieves similar processing speeds
Relators
Academic year
Publication type
Number of Pages
Course of studies
Classe di laurea
Ente in cotutela
Aziende collaboratrici
URI
![]() |
Modify record (reserved for operators) |
