1. Precise Discrimination for Multiple Underlying Pathologies of Dementia Cases Based on Deep-Learning with Electroencephalography
- Author
-
Tatsuya Harada, Yusuke Watanabe, Shunichiro Ikeda, Kenji Yoshiyama, Hiroaki Kazui, Yoshiteru Takekita, Kimihisa Awata, Masahiro Hata, Yuto Satake, Hideki Kanemoto, Ryohei Fukuma, Keiichiro Nishida, Manabu Ikeda, Yuki Miyazaki, Takumi Tanaka, Daiki Taomoto, Ryouhei Ishii, Masao Iwase, Takashi Suehiro, Takufumi Yanagisawa, Masafumi Yoshimura, and Haruhiko Kishima
- Subjects
Text mining ,medicine.diagnostic_test ,business.industry ,Deep learning ,mental disorders ,medicine ,Dementia ,Artificial intelligence ,Electroencephalography ,medicine.disease ,Psychology ,business ,Cognitive psychology - Abstract
Background: Developing accurate and universally available biomarkers for dementia diseases is demanded under world-wide rapid increasing of patients with dementia. Electroencephalogram (EEG) offers promising examinations due to their inexpensiveness, high availability, and sensitiveness to neural functions. EEG applicability can be expanded by deep-learning.Methods: We analyzed EEG signals based on novel deep neural network in healthy volunteers (HV, N=55), patients with Alzheimer's disease (AD, N=101), dementia with Lewy bodies (DLB, N=75), and idiopathic normal pressure hydrocephalus (iNPH, N=60) to evaluate the discriminative accuracy of these diseases.Results: High discriminative accuracies were archived between HV and patients with dementia, yielding 81.7 %(vs AD), 93.9% (vs DLB), and 93.1% (vs iNPH).Conclusions: This study revealed that the EEG data of patients with dementia were successfully discriminated from healthy volunteers based on deep learning and could produce a new purpose of EEG measurement in screening for dementia diseases.
- Published
- 2021
- Full Text
- View/download PDF