OBJECTIVE: Key issues in the epilepsy seizure prediction research are (1) the reproducibility of results (2) the inability to compare multiple approaches directly. To overcome these problems, the seizure prediction challenge was organized on Kaggle.com. It aimed at establishing benchmarks on a dataset with predefined train, validation, and test sets. Our main objective is to analyze the competition format, and to propose improvements, which would facilitate a better comparison of algorithms. The second objective is to present a novel deep learning approach to seizure prediction and compare it to other commonly used methods using patient centered metrics. METHODS: We used the competition's datasets to illustrate the effects of data contamination. Having better data partitions, we compared three types of models in terms of different objectives. RESULTS: We found that correct selection of test samples is crucial when evaluating the performance of seizure forecasting models. Moreover, we showed that models, which achieve state-of-the-art performance with respect to commonly used AUC, sensitivity, and specificity metrics, may not yet be suitable for practical usage because of low precision scores. CONCLUSION: Correlation between validation and test datasets used in the competition limited its scientific value. SIGNIFICANCE: Our findings provide guidelines which allow for a more objective evaluation of seizure prediction models.
OBJECTIVE: Key issues in the epilepsy seizure prediction research are (1) the reproducibility of results (2) the inability to compare multiple approaches directly. To overcome these problems, the seizure prediction challenge was organized on Kaggle.com. It aimed at establishing benchmarks on a dataset with predefined train, validation, and test sets. Our main objective is to analyze the competition format, and to propose improvements, which would facilitate a better comparison of algorithms. The second objective is to present a novel deep learning approach to seizure prediction and compare it to other commonly used methods using patient centered metrics. METHODS: We used the competition's datasets to illustrate the effects of data contamination. Having better data partitions, we compared three types of models in terms of different objectives. RESULTS: We found that correct selection of test samples is crucial when evaluating the performance of seizure forecasting models. Moreover, we showed that models, which achieve state-of-the-art performance with respect to commonly used AUC, sensitivity, and specificity metrics, may not yet be suitable for practical usage because of low precision scores. CONCLUSION: Correlation between validation and test datasets used in the competition limited its scientific value. SIGNIFICANCE: Our findings provide guidelines which allow for a more objective evaluation of seizure prediction models.
Authors: Mona Nasseri; Vaclav Kremen; Petr Nejedly; Inyong Kim; Su-Youne Chang; Hang Joon Jo; Hari Guragain; Nathaniel Nelson; Edward Patterson; Beverly K Sturges; Chelsea M Crowe; Tim Denison; Benjamin H Brinkmann; Gregory A Worrell Journal: Biomed Signal Process Control Date: 2019-11-14 Impact factor: 3.880
Authors: Benjamin H Brinkmann; Philippa J Karoly; Ewan S Nurse; Sonya B Dumanis; Mona Nasseri; Pedro F Viana; Andreas Schulze-Bonhage; Dean R Freestone; Greg Worrell; Mark P Richardson; Mark J Cook Journal: Front Neurol Date: 2021-07-13 Impact factor: 4.003