Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset
For automatic EEG diagnosis, this paper presents a new EEG data set with well-organized clinical annotations called Chung-Ang University Hospital EEG (CAUEEG), which has event history, patient’s age, and corresponding diagnosis labels. We also designed two reliable evaluation tasks for the low-cost,...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-05-01
|
Series: | NeuroImage |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S1053811923002008 |
_version_ | 1797847709674110976 |
---|---|
author | Min-jae Kim Young Chul Youn Joonki Paik |
author_facet | Min-jae Kim Young Chul Youn Joonki Paik |
author_sort | Min-jae Kim |
collection | DOAJ |
description | For automatic EEG diagnosis, this paper presents a new EEG data set with well-organized clinical annotations called Chung-Ang University Hospital EEG (CAUEEG), which has event history, patient’s age, and corresponding diagnosis labels. We also designed two reliable evaluation tasks for the low-cost, non-invasive diagnosis to detect brain disorders: i) CAUEEG-Dementia with normal, mci, and dementia diagnostic labels and ii) CAUEEG-Abnormal with normal and abnormal. Based on the CAUEEG dataset, this paper proposes a new fully end-to-end deep learning model, called the CAUEEG End-to-end Deep neural Network (CEEDNet). CEEDNet pursues to bring all the functional elements for the EEG analysis in a seamless learnable fashion while restraining non-essential human intervention. Extensive experiments showed that our CEEDNet significantly improves the accuracy compared with existing methods, such as machine learning methods and Ieracitano-CNN (Ieracitano et al., 2019), due to taking full advantage of end-to-end learning. The high ROC-AUC scores of 0.9 on CAUEEG-Dementia and 0.86 on CAUEEG-Abnormal recorded by our CEEDNet models demonstrate that our method can lead potential patients to early diagnosis through automatic screening. |
first_indexed | 2024-04-09T18:16:44Z |
format | Article |
id | doaj.art-83e69e1af6f94f3fbb9f9d52689812db |
institution | Directory Open Access Journal |
issn | 1095-9572 |
language | English |
last_indexed | 2024-04-09T18:16:44Z |
publishDate | 2023-05-01 |
publisher | Elsevier |
record_format | Article |
series | NeuroImage |
spelling | doaj.art-83e69e1af6f94f3fbb9f9d52689812db2023-04-13T04:26:06ZengElsevierNeuroImage1095-95722023-05-01272120054Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and datasetMin-jae Kim0Young Chul Youn1Joonki Paik2Department of Image, Chung-Ang University, Seoul, 06974, South KoreaDepartment of Neurology, Chung-Ang University College of Medicine, Seoul, 06973, South Korea; Biomedical Research Institute, Chung-Ang University Hospital, Seoul, 06973, South KoreaCorresponding author.; Department of Image, Chung-Ang University, Seoul, 06974, South Korea; Department of Artificial Intelligence, Chung-Ang University, Seoul, 06974, South KoreaFor automatic EEG diagnosis, this paper presents a new EEG data set with well-organized clinical annotations called Chung-Ang University Hospital EEG (CAUEEG), which has event history, patient’s age, and corresponding diagnosis labels. We also designed two reliable evaluation tasks for the low-cost, non-invasive diagnosis to detect brain disorders: i) CAUEEG-Dementia with normal, mci, and dementia diagnostic labels and ii) CAUEEG-Abnormal with normal and abnormal. Based on the CAUEEG dataset, this paper proposes a new fully end-to-end deep learning model, called the CAUEEG End-to-end Deep neural Network (CEEDNet). CEEDNet pursues to bring all the functional elements for the EEG analysis in a seamless learnable fashion while restraining non-essential human intervention. Extensive experiments showed that our CEEDNet significantly improves the accuracy compared with existing methods, such as machine learning methods and Ieracitano-CNN (Ieracitano et al., 2019), due to taking full advantage of end-to-end learning. The high ROC-AUC scores of 0.9 on CAUEEG-Dementia and 0.86 on CAUEEG-Abnormal recorded by our CEEDNet models demonstrate that our method can lead potential patients to early diagnosis through automatic screening.http://www.sciencedirect.com/science/article/pii/S1053811923002008Electroencephalography (EEG)DementiaMild cognitive impairment (MCI)Deep learningConvolutional neural network (CNN)Automatic diagnostic system |
spellingShingle | Min-jae Kim Young Chul Youn Joonki Paik Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset NeuroImage Electroencephalography (EEG) Dementia Mild cognitive impairment (MCI) Deep learning Convolutional neural network (CNN) Automatic diagnostic system |
title | Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset |
title_full | Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset |
title_fullStr | Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset |
title_full_unstemmed | Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset |
title_short | Deep learning-based EEG analysis to classify normal, mild cognitive impairment, and dementia: Algorithms and dataset |
title_sort | deep learning based eeg analysis to classify normal mild cognitive impairment and dementia algorithms and dataset |
topic | Electroencephalography (EEG) Dementia Mild cognitive impairment (MCI) Deep learning Convolutional neural network (CNN) Automatic diagnostic system |
url | http://www.sciencedirect.com/science/article/pii/S1053811923002008 |
work_keys_str_mv | AT minjaekim deeplearningbasedeeganalysistoclassifynormalmildcognitiveimpairmentanddementiaalgorithmsanddataset AT youngchulyoun deeplearningbasedeeganalysistoclassifynormalmildcognitiveimpairmentanddementiaalgorithmsanddataset AT joonkipaik deeplearningbasedeeganalysistoclassifynormalmildcognitiveimpairmentanddementiaalgorithmsanddataset |