2022
DOI: 10.1016/j.seizure.2022.01.009
|View full text |Cite
|
Sign up to set email alerts
|

Automated seizure detection in an EMU setting: Are software packages ready for implementation?

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
25
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
1
1

Relationship

1
6

Authors

Journals

citations
Cited by 13 publications
(27 citation statements)
references
References 17 publications
2
25
0
Order By: Relevance
“…Commercial seizure detectors can assist clinicians in reviewing EEG recordings, which may help to reduce time and boost the annotation accuracy 53 . Earlier studies by Reus et al 16 and Koren et al 17 have compared the performance of Persyst, Encevis, and BESA. We summarized their őndings against the performance of the proposed detector in Table 6.…”
Section: Comparison With Commercial Seizure Detectorsmentioning
confidence: 99%
See 3 more Smart Citations
“…Commercial seizure detectors can assist clinicians in reviewing EEG recordings, which may help to reduce time and boost the annotation accuracy 53 . Earlier studies by Reus et al 16 and Koren et al 17 have compared the performance of Persyst, Encevis, and BESA. We summarized their őndings against the performance of the proposed detector in Table 6.…”
Section: Comparison With Commercial Seizure Detectorsmentioning
confidence: 99%
“…After collecting all the start and end times of the detected seizures, we assess the accuracy of those detections. There are several well-established evaluation metrics, such as epoch-based sampling (EBS) 46 , any-overlap (OVLP) 46 , time-aligned event scoring (TAES) 46 , and increased margin scoring (IMS) 16 . However, these metrics do not accurately reŕect the requirement of a seizure detector deployed in clinical settings.…”
Section: Eeg-level Seizure Detection Evaluation Metricmentioning
confidence: 99%
See 2 more Smart Citations
“…To eliminate ambiguity, one must adequately define the evaluation metric to compute the event-based TP, FN, and FP (TN is not computed for EEG-level seizure detection). There are several well-established evaluation metrics, such as epoch-based sampling (EBS) [242], any-overlap (OVLP) [242], increased margin scoring (IMS) [243], and time-aligned event scoring (TAES) [242]. Unfortunately, the seizure evaluation metric is often not reported.…”
Section: Expected Calibration Error (Ece)mentioning
confidence: 99%