Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks

Understanding brain function relies on identifying spatiotemporal patterns in brain activity. In recent years, machine learning methods have been widely used to detect connections between regions of interest (ROIs) involved in cognitive functions, as measured by the fMRI technique. However, it'...

Full description

Saved in:
Bibliographic Details
Main Authors: Magdalena Fafrowicz, Marcin Tutajewski, Igor Sieradzki, Jeremi K. Ochab, Anna Ceglarek-Sroka, Koryna Lewandowska, Tadeusz Marek, Barbara Sikora-Wachowicz, Igor T. Podolak, Paweł Oświęcimka
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-12-01
Series:Frontiers in Neuroinformatics
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fninf.2024.1480366/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850132335734816768
author Magdalena Fafrowicz
Marcin Tutajewski
Igor Sieradzki
Jeremi K. Ochab
Jeremi K. Ochab
Anna Ceglarek-Sroka
Koryna Lewandowska
Tadeusz Marek
Barbara Sikora-Wachowicz
Igor T. Podolak
Paweł Oświęcimka
Paweł Oświęcimka
Paweł Oświęcimka
author_facet Magdalena Fafrowicz
Marcin Tutajewski
Igor Sieradzki
Jeremi K. Ochab
Jeremi K. Ochab
Anna Ceglarek-Sroka
Koryna Lewandowska
Tadeusz Marek
Barbara Sikora-Wachowicz
Igor T. Podolak
Paweł Oświęcimka
Paweł Oświęcimka
Paweł Oświęcimka
author_sort Magdalena Fafrowicz
collection DOAJ
description Understanding brain function relies on identifying spatiotemporal patterns in brain activity. In recent years, machine learning methods have been widely used to detect connections between regions of interest (ROIs) involved in cognitive functions, as measured by the fMRI technique. However, it's essential to match the type of learning method to the problem type, and extracting the information about the most important ROI connections might be challenging. In this contribution, we used machine learning techniques to classify tasks in a working memory experiment and identify the brain areas involved in processing information. We employed classical discriminators and neural networks (convolutional and residual) to differentiate between brain responses to distinct types of visual stimuli (visuospatial and verbal) and different phases of the experiment (information encoding and retrieval). The best performance was achieved by the LGBM classifier with 1-time point input data during memory retrieval and a convolutional neural network during the encoding phase. Additionally, we developed an algorithm that took into account feature correlations to estimate the most important brain regions for the model's accuracy. Our findings suggest that from the perspective of considered models, brain signals related to the resting state have a similar degree of complexity to those related to the encoding phase, which does not improve the model's accuracy. However, during the retrieval phase, the signals were easily distinguished from the resting state, indicating their different structure. The study identified brain regions that are crucial for processing information in working memory, as well as the differences in the dynamics of encoding and retrieval processes. Furthermore, our findings indicate spatiotemporal distinctions related to these processes. The analysis confirmed the importance of the basal ganglia in processing information during the retrieval phase. The presented results reveal the benefits of applying machine learning algorithms to investigate working memory dynamics.
format Article
id doaj-art-23a2df9ccea449cda0c0479acdfbee5e
institution OA Journals
issn 1662-5196
language English
publishDate 2024-12-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neuroinformatics
spelling doaj-art-23a2df9ccea449cda0c0479acdfbee5e2025-08-20T02:32:14ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962024-12-011810.3389/fninf.2024.14803661480366Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networksMagdalena Fafrowicz0Marcin Tutajewski1Igor Sieradzki2Jeremi K. Ochab3Jeremi K. Ochab4Anna Ceglarek-Sroka5Koryna Lewandowska6Tadeusz Marek7Barbara Sikora-Wachowicz8Igor T. Podolak9Paweł Oświęcimka10Paweł Oświęcimka11Paweł Oświęcimka12Department of Cognitive Neuroscience and Neuroergonomics, Jagiellonian University, Kraków, PolandInstitute of Theoretical Physics, Jagiellonian University, Kraków, PolandGroup of Machine Learning Methods GMUM, Faculty of Mathematics and Computer Science, Jagiellonian University, Kraków, PolandInstitute of Theoretical Physics, Jagiellonian University, Kraków, PolandMark Kac Center for Complex Systems Research, Jagiellonian University, Kraków, PolandDepartment of Cognitive Neuroscience and Neuroergonomics, Jagiellonian University, Kraków, PolandDepartment of Cognitive Neuroscience and Neuroergonomics, Jagiellonian University, Kraków, PolandFaculty of Psychology, SWPS University, Katowice, PolandDepartment of Cognitive Neuroscience and Neuroergonomics, Jagiellonian University, Kraków, PolandGroup of Machine Learning Methods GMUM, Faculty of Mathematics and Computer Science, Jagiellonian University, Kraków, PolandInstitute of Theoretical Physics, Jagiellonian University, Kraków, PolandGroup of Machine Learning Methods GMUM, Faculty of Mathematics and Computer Science, Jagiellonian University, Kraków, PolandComplex Systems Theory Department, Institute of Nuclear Physics, Polish Academy of Sciences, Kraków, PolandUnderstanding brain function relies on identifying spatiotemporal patterns in brain activity. In recent years, machine learning methods have been widely used to detect connections between regions of interest (ROIs) involved in cognitive functions, as measured by the fMRI technique. However, it's essential to match the type of learning method to the problem type, and extracting the information about the most important ROI connections might be challenging. In this contribution, we used machine learning techniques to classify tasks in a working memory experiment and identify the brain areas involved in processing information. We employed classical discriminators and neural networks (convolutional and residual) to differentiate between brain responses to distinct types of visual stimuli (visuospatial and verbal) and different phases of the experiment (information encoding and retrieval). The best performance was achieved by the LGBM classifier with 1-time point input data during memory retrieval and a convolutional neural network during the encoding phase. Additionally, we developed an algorithm that took into account feature correlations to estimate the most important brain regions for the model's accuracy. Our findings suggest that from the perspective of considered models, brain signals related to the resting state have a similar degree of complexity to those related to the encoding phase, which does not improve the model's accuracy. However, during the retrieval phase, the signals were easily distinguished from the resting state, indicating their different structure. The study identified brain regions that are crucial for processing information in working memory, as well as the differences in the dynamics of encoding and retrieval processes. Furthermore, our findings indicate spatiotemporal distinctions related to these processes. The analysis confirmed the importance of the basal ganglia in processing information during the retrieval phase. The presented results reveal the benefits of applying machine learning algorithms to investigate working memory dynamics.https://www.frontiersin.org/articles/10.3389/fninf.2024.1480366/fullexplainabilityfMRIworking memoryROImachine learningneural network
spellingShingle Magdalena Fafrowicz
Marcin Tutajewski
Igor Sieradzki
Jeremi K. Ochab
Jeremi K. Ochab
Anna Ceglarek-Sroka
Koryna Lewandowska
Tadeusz Marek
Barbara Sikora-Wachowicz
Igor T. Podolak
Paweł Oświęcimka
Paweł Oświęcimka
Paweł Oświęcimka
Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
Frontiers in Neuroinformatics
explainability
fMRI
working memory
ROI
machine learning
neural network
title Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
title_full Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
title_fullStr Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
title_full_unstemmed Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
title_short Classification of ROI-based fMRI data in short-term memory tasks using discriminant analysis and neural networks
title_sort classification of roi based fmri data in short term memory tasks using discriminant analysis and neural networks
topic explainability
fMRI
working memory
ROI
machine learning
neural network
url https://www.frontiersin.org/articles/10.3389/fninf.2024.1480366/full
work_keys_str_mv AT magdalenafafrowicz classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT marcintutajewski classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT igorsieradzki classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT jeremikochab classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT jeremikochab classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT annaceglareksroka classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT korynalewandowska classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT tadeuszmarek classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT barbarasikorawachowicz classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT igortpodolak classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT pawełoswiecimka classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT pawełoswiecimka classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks
AT pawełoswiecimka classificationofroibasedfmridatainshorttermmemorytasksusingdiscriminantanalysisandneuralnetworks