Amplifying pathological detection in EEG signaling pathways through cross-dataset transfer learning.

Autor: Darvishi-Bayazi MJ; Mila, Québec AI Institute, Montréal, QC, Canada; Faubert Lab, Montréal, QC, Canada; Université de Montréal, Montréal, QC, Canada. Electronic address: mohammad.bayazi@mila.quebec., Ghaemi MS; National Research Council Canada, Toronto, ON, Canada., Lesort T; Mila, Québec AI Institute, Montréal, QC, Canada; Université de Montréal, Montréal, QC, Canada., Arefin MR; Mila, Québec AI Institute, Montréal, QC, Canada; Université de Montréal, Montréal, QC, Canada., Faubert J; Faubert Lab, Montréal, QC, Canada; Université de Montréal, Montréal, QC, Canada., Rish I; Mila, Québec AI Institute, Montréal, QC, Canada; Université de Montréal, Montréal, QC, Canada.
Jazyk: angličtina
Zdroj: Computers in biology and medicine [Comput Biol Med] 2024 Feb; Vol. 169, pp. 107893. Date of Electronic Publication: 2023 Dec 30.
DOI: 10.1016/j.compbiomed.2023.107893
Abstrakt: Pathology diagnosis based on EEG signals and decoding brain activity holds immense importance in understanding neurological disorders. With the advancement of artificial intelligence methods and machine learning techniques, the potential for accurate data-driven diagnoses and effective treatments has grown significantly. However, applying machine learning algorithms to real-world datasets presents diverse challenges at multiple levels. The scarcity of labeled data, especially in low regime scenarios with limited availability of real patient cohorts due to high costs of recruitment, underscores the vital deployment of scaling and transfer learning techniques. In this study, we explore a real-world pathology classification task to highlight the effectiveness of data and model scaling and cross-dataset knowledge transfer. As such, we observe varying performance improvements through data scaling, indicating the need for careful evaluation and labeling. Additionally, we identify the challenges of possible negative transfer and emphasize the significance of some key components to overcome distribution shifts and potential spurious correlations and achieve positive transfer. We see improvement in the performance of the target model on the target (NMT) datasets by using the knowledge from the source dataset (TUAB) when a low amount of labeled data was available. Our findings demonstrated that a small and generic model (e.g. ShallowNet) performs well on a single dataset, however, a larger model (e.g. TCN) performs better in transfer learning when leveraging a larger and more diverse dataset.
Competing Interests: Declaration of competing interest None Declared.
(Copyright © 2023 The Author(s). Published by Elsevier Ltd.. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: