Nitin K.Sethi, Associate Professor of Neurology, New York-Presbyterian Hospital, Weill Cornell Medical Center (New York, NY)
Submitted March 08, 2019
I read with interest González Otárula et al.’s study on automated seizure detection accuracy for ambulatory EEG recordings and the accompanying editorial by Juhasz and Berg.1,2 The authors found disappointing accuracy of currently available automated methods of ictal event detection in their selected samples of aEEG. Artificial intelligence, machine learning, and deep learning are the tools that likely shall provide algorithms that shall improve the current dismal rates of automated methods of ictal event detection. The technology is already present; what we are lacking is funding, multi-institutional collaboration, and a large repository of EEG samples available to researchers who are working on this issue. If we can teach a car to drive itself, we can certainly teach a CT scan or an EEG to read itself.
References
González Otárula KA, Mikhaeil-Demo Y, Bachman EM, Balaguera P, Schuele S. Automated seizure detection accuracy for ambulatory EEG recordings. Neurology 2019;92:e5140–e1546.
Juhász C, Berg M. Computerized seizure detection on ambulatory EEG: Finding the needles in the haystack. Neurology 2019;92:641–642.
I read with interest González Otárula et al.’s study on automated seizure detection accuracy for ambulatory EEG recordings and the accompanying editorial by Juhasz and Berg.1,2 The authors found disappointing accuracy of currently available automated methods of ictal event detection in their selected samples of aEEG. Artificial intelligence, machine learning, and deep learning are the tools that likely shall provide algorithms that shall improve the current dismal rates of automated methods of ictal event detection. The technology is already present; what we are lacking is funding, multi-institutional collaboration, and a large repository of EEG samples available to researchers who are working on this issue. If we can teach a car to drive itself, we can certainly teach a CT scan or an EEG to read itself.
References
Footnote
For disclosures, please contact the editorial office at journal@neurology.org.