EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Methodological Issues in Evaluating Machine Learning Models for EEG Seizure Prediction: Good Cross-Validation Accuracy Does Not Guarantee Generalization to New Patients

Sina Shafiezadeh, Gian Marco Duma, Giovanni Mento, Alberto Danieli, Lisa Antoniazzi, Fiorella Del Popolo Cristaldi, Paolo Bonanni, Alberto Testolin

Applied Sciences · 28.03.2023 · 13 (7) · 4262

DOI: 10.3390/app13074262 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 28.03.2023. Verfasst von Sina Shafiezadeh und 7 weiteren Autor:innen. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Italian Health Ministry.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

There is an increasing interest in applying artificial intelligence techniques to forecast epileptic seizures. In particular, machine learning algorithms could extract nonlinear statistical regularities from electroencephalographic (EEG) time series that can anticipate abnormal brain activity. The recent literature reports promising results in seizure detection and prediction tasks using machine and deep learning methods. However, performance evaluation is often based on questionable randomized cross-validation schemes, which can introduce correlated signals (e.g., EEG data recorded from the same patient during nearby periods of the day) into the partitioning of training and test sets. The present study demonstrates that the use of more stringent evaluation strategies, such as those based on leave-one-patient-out partitioning, leads to a drop in accuracy from about 80% to 50% for a standard eXtreme Gradient Boosting (XGBoost) classifier on two different data sets. Our findings suggest that the definition of rigorous evaluation protocols is crucial to ensure the generalizability of predictive models before proceeding to clinical trials.

Förderung

  • Italian Health Ministry

Zitieren

APA

Shafiezadeh, S., Duma, G. M., Mento, G., Danieli, A., Antoniazzi, L., Cristaldi, F. D. P., Bonanni, P., & Testolin, A. (2023). Methodological Issues in Evaluating Machine Learning Models for EEG Seizure Prediction: Good Cross-Validation Accuracy Does Not Guarantee Generalization to New Patients. Applied Sciences, 13(7), 4262. https://doi.org/10.3390/app13074262

Vancouver

Shafiezadeh S, Duma GM, Mento G, Danieli A, Antoniazzi L, Cristaldi FDP, et al. Methodological Issues in Evaluating Machine Learning Models for EEG Seizure Prediction: Good Cross-Validation Accuracy Does Not Guarantee Generalization to New Patients. Applied Sciences. 2023;13(7):4262. doi:10.3390/app13074262

BibTeX
@article{Shafiezadeh2023,
  author = {Shafiezadeh, Sina and Duma, Gian Marco and Mento, Giovanni and Danieli, Alberto and Antoniazzi, Lisa and Cristaldi, Fiorella Del Popolo and Bonanni, Paolo and Testolin, Alberto},
  title = {Methodological Issues in Evaluating Machine Learning Models for EEG Seizure Prediction: Good Cross-Validation Accuracy Does Not Guarantee Generalization to New Patients},
  journal = {Applied Sciences},
  year = {2023},
  volume = {13},
  number = {7},
  pages = {4262},
  doi = {10.3390/app13074262}
}
Kostenlos starten →