The Choice of Evaluation Metrics in the Prediction of Epileptiform Activity

会议论文
发表日期：

2024年

查看全文

Nikolay Gromov,
Albina Lebedeva,
Ivan Kipelkin,
Oksana Elshina,
Konstantin Yashin,
Lev Smirnov,
Tatiana Levanova,
Svetlana Gerasimova

摘要

In this study, we investigate the problem of prediction of epileptiform activity from EEG data using a deep learning approach. We implement LSTM deep neural network and study how the quality of the prediction depends on the choice of measures of observational error, such as MAPE and RMSE. We show that comparison of results obtained using different metrics is important to obtain a comprehensive assessment of the problem.