Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes
ObjectivesMore than half of patients with acute ischemic stroke develop post-stroke cognitive impairment (PSCI), a significant barrier to future neurological recovery. Thus, predicting cognitive trajectories post-AIS is crucial. Our primary objective is to determine whether brain network properties...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-09-01
|
| Series: | Frontiers in Aging Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnagi.2023.1238274/full |
| _version_ | 1797671457551024128 |
|---|---|
| author | Minwoo Lee Yuseong Hong Sungsik An Ukeob Park Jaekang Shin Jeongjae Lee Mi Sun Oh Byung-Chul Lee Kyung-Ho Yu Jae-Sung Lim Seung Wan Kang |
| author_facet | Minwoo Lee Yuseong Hong Sungsik An Ukeob Park Jaekang Shin Jeongjae Lee Mi Sun Oh Byung-Chul Lee Kyung-Ho Yu Jae-Sung Lim Seung Wan Kang |
| author_sort | Minwoo Lee |
| collection | DOAJ |
| description | ObjectivesMore than half of patients with acute ischemic stroke develop post-stroke cognitive impairment (PSCI), a significant barrier to future neurological recovery. Thus, predicting cognitive trajectories post-AIS is crucial. Our primary objective is to determine whether brain network properties from electroencephalography (EEG) can predict post-stroke cognitive function using machine learning approach.MethodsWe enrolled consecutive stroke patients who underwent both EEG during the acute stroke phase and cognitive assessments 3 months post-stroke. We preprocessed acute stroke EEG data to eliminate low-quality epochs, then performed independent component analysis and quantified network characteristics using iSyncBrain®. Cognitive function was evaluated using the Montreal cognitive assessment (MoCA). We initially categorized participants based on the lateralization of their lesions and then developed machine learning models to predict cognitive status in the left and right hemisphere lesion groups.ResultsEighty-seven patients were included, and the accuracy of lesion laterality prediction using EEG attributes was 97.0%. In the left hemispheric lesion group, the network attributes of the theta band were significantly correlated with MoCA scores, and higher global efficiency, clustering coefficient, and lower characteristic path length were associated with higher MoCA scores. Most features related to cognitive scores were selected from the frontal lobe. The predictive powers (R-squared) were 0.76 and 0.65 for the left and right stroke groups, respectively.ConclusionEstimating EEG-based network properties in the acute phase of ischemic stroke through a machine learning model has a potential to predict cognitive outcomes after ischemic stroke. |
| first_indexed | 2024-03-11T21:15:50Z |
| format | Article |
| id | doaj.art-4e20795074c34501b140ce4407880f2e |
| institution | Directory Open Access Journal |
| issn | 1663-4365 |
| language | English |
| last_indexed | 2024-03-11T21:15:50Z |
| publishDate | 2023-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Aging Neuroscience |
| spelling | doaj.art-4e20795074c34501b140ce4407880f2e2023-09-29T03:59:28ZengFrontiers Media S.A.Frontiers in Aging Neuroscience1663-43652023-09-011510.3389/fnagi.2023.12382741238274Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributesMinwoo Lee0Yuseong Hong1Sungsik An2Ukeob Park3Jaekang Shin4Jeongjae Lee5Mi Sun Oh6Byung-Chul Lee7Kyung-Ho Yu8Jae-Sung Lim9Seung Wan Kang10Department of Neurology, Hallym University Sacred Heart Hospital, Hallym Neurological Institute, Hallym University College of Medicine, Anyang, Republic of KoreaiMedisync, Inc., Seoul, Republic of KoreaDepartment of Neurology, Hwahong Hospital, Suwon, Republic of KoreaiMedisync, Inc., Seoul, Republic of KoreaiMedisync, Inc., Seoul, Republic of KoreaDepartment of Neurology, Hallym University Sacred Heart Hospital, Hallym Neurological Institute, Hallym University College of Medicine, Anyang, Republic of KoreaDepartment of Neurology, Hallym University Sacred Heart Hospital, Hallym Neurological Institute, Hallym University College of Medicine, Anyang, Republic of KoreaDepartment of Neurology, Hallym University Sacred Heart Hospital, Hallym Neurological Institute, Hallym University College of Medicine, Anyang, Republic of KoreaDepartment of Neurology, Hallym University Sacred Heart Hospital, Hallym Neurological Institute, Hallym University College of Medicine, Anyang, Republic of KoreaDepartment of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of KoreaiMedisync, Inc., Seoul, Republic of KoreaObjectivesMore than half of patients with acute ischemic stroke develop post-stroke cognitive impairment (PSCI), a significant barrier to future neurological recovery. Thus, predicting cognitive trajectories post-AIS is crucial. Our primary objective is to determine whether brain network properties from electroencephalography (EEG) can predict post-stroke cognitive function using machine learning approach.MethodsWe enrolled consecutive stroke patients who underwent both EEG during the acute stroke phase and cognitive assessments 3 months post-stroke. We preprocessed acute stroke EEG data to eliminate low-quality epochs, then performed independent component analysis and quantified network characteristics using iSyncBrain®. Cognitive function was evaluated using the Montreal cognitive assessment (MoCA). We initially categorized participants based on the lateralization of their lesions and then developed machine learning models to predict cognitive status in the left and right hemisphere lesion groups.ResultsEighty-seven patients were included, and the accuracy of lesion laterality prediction using EEG attributes was 97.0%. In the left hemispheric lesion group, the network attributes of the theta band were significantly correlated with MoCA scores, and higher global efficiency, clustering coefficient, and lower characteristic path length were associated with higher MoCA scores. Most features related to cognitive scores were selected from the frontal lobe. The predictive powers (R-squared) were 0.76 and 0.65 for the left and right stroke groups, respectively.ConclusionEstimating EEG-based network properties in the acute phase of ischemic stroke through a machine learning model has a potential to predict cognitive outcomes after ischemic stroke.https://www.frontiersin.org/articles/10.3389/fnagi.2023.1238274/fullischemic strokecognitionelectroencephalographyfunctional networkmachine learning |
| spellingShingle | Minwoo Lee Yuseong Hong Sungsik An Ukeob Park Jaekang Shin Jeongjae Lee Mi Sun Oh Byung-Chul Lee Kyung-Ho Yu Jae-Sung Lim Seung Wan Kang Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes Frontiers in Aging Neuroscience ischemic stroke cognition electroencephalography functional network machine learning |
| title | Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes |
| title_full | Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes |
| title_fullStr | Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes |
| title_full_unstemmed | Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes |
| title_short | Machine learning-based prediction of post-stroke cognitive status using electroencephalography-derived brain network attributes |
| title_sort | machine learning based prediction of post stroke cognitive status using electroencephalography derived brain network attributes |
| topic | ischemic stroke cognition electroencephalography functional network machine learning |
| url | https://www.frontiersin.org/articles/10.3389/fnagi.2023.1238274/full |
| work_keys_str_mv | AT minwoolee machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT yuseonghong machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT sungsikan machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT ukeobpark machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT jaekangshin machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT jeongjaelee machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT misunoh machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT byungchullee machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT kyunghoyu machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT jaesunglim machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes AT seungwankang machinelearningbasedpredictionofpoststrokecognitivestatususingelectroencephalographyderivedbrainnetworkattributes |