A deep learning based framework for diagnosis of mild cognitive impairment
作者：
Highlights：
• For the first time, an LSTM and GRU-based deep learning study for MCI classification has been accomplished.
• We have utilized the LSTM-extracted features for SVM and KNN instead of using separate feature extraction methods to save computational cost.
• To enhance the suggested model’s performance, we have examined the average filtering strategy for down sampling.
• With reduced calculation time, the consistency of the performance has been checked by 5-fold cross validation scheme.
摘要
•For the first time, an LSTM and GRU-based deep learning study for MCI classification has been accomplished.•We have utilized the LSTM-extracted features for SVM and KNN instead of using separate feature extraction methods to save computational cost.•To enhance the suggested model’s performance, we have examined the average filtering strategy for down sampling.•With reduced calculation time, the consistency of the performance has been checked by 5-fold cross validation scheme.
论文关键词：Mild cognitive impairment (MCI),Electroencephalogram (EEG),Alzheimer’s diseases (AD),Long Short-Term Memory (LSTM),Gated Recurrent Unit (GRU)
论文评审过程：Received 23 February 2022, Revised 4 April 2022, Accepted 12 April 2022, Available online 25 April 2022, Version of Record 11 May 2022.
论文官网地址：https://doi.org/10.1016/j.knosys.2022.108815
