Guardado en:
Detalles Bibliográficos
Autores principales: Aggarwal, Himanshu, Al-Shikhley, Liza, Thirion, Bertrand
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2407.12056
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909466124877824
author Aggarwal, Himanshu
Al-Shikhley, Liza
Thirion, Bertrand
author_facet Aggarwal, Himanshu
Al-Shikhley, Liza
Thirion, Bertrand
contents Decoding cognitive states from functional magnetic resonance imaging is central to understanding the functional organization of the brain. Within-subject decoding avoids between-subject correspondence problems but requires large sample sizes to make accurate predictions; obtaining such large sample sizes is both challenging and expensive. Here, we investigate an ensemble approach to decoding that combines the classifiers trained on data from other subjects to decode cognitive states in a new subject. We compare it with the conventional decoding approach on five different datasets and cognitive tasks. We find that it outperforms the conventional approach by up to 20% in accuracy, especially for datasets with limited per-subject data. The ensemble approach is particularly advantageous when the classifier is trained in voxel space. Furthermore, a Multi-layer Perceptron turns out to be a good default choice as an ensemble method. These results show that the pre-training strategy reduces the need for large per-subject data.
format Preprint
id arxiv_https___arxiv_org_abs_2407_12056
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Across-subject ensemble-learning alleviates the need for large samples for fMRI decoding
Aggarwal, Himanshu
Al-Shikhley, Liza
Thirion, Bertrand
Image and Video Processing
Machine Learning
Decoding cognitive states from functional magnetic resonance imaging is central to understanding the functional organization of the brain. Within-subject decoding avoids between-subject correspondence problems but requires large sample sizes to make accurate predictions; obtaining such large sample sizes is both challenging and expensive. Here, we investigate an ensemble approach to decoding that combines the classifiers trained on data from other subjects to decode cognitive states in a new subject. We compare it with the conventional decoding approach on five different datasets and cognitive tasks. We find that it outperforms the conventional approach by up to 20% in accuracy, especially for datasets with limited per-subject data. The ensemble approach is particularly advantageous when the classifier is trained in voxel space. Furthermore, a Multi-layer Perceptron turns out to be a good default choice as an ensemble method. These results show that the pre-training strategy reduces the need for large per-subject data.
title Across-subject ensemble-learning alleviates the need for large samples for fMRI decoding
topic Image and Video Processing
Machine Learning
url https://arxiv.org/abs/2407.12056