OBJECTIVE: High frequency oscillations (HFOs) are a biomarker of epileptogenicity. Visual marking of HFOs is highly time-consuming and inevitably subjective, making automatic detection necessary. We compare four existing detectors on the same dataset. METHODS: HFOs and baselines were identified by experienced reviewers in intracerebral EEGs from 20 patients. A new feature of our detector to deal with channels where baseline cannot be found is presented. The original and an optimal configuration are implemented. Receiver operator curves, false discovery rate, and channel ranking are used to evaluate performance. RESULTS: All detectors improve performance with the optimal configuration. Our detector had higher sensitivity, lower false positives than the others, and similar false detections. The main difference in performance was in very active channels. CONCLUSIONS: Each detector was developed for different recordings and with different aims. Our detector performed better in this dataset, but was developed on data similar to the test data. Moreover, optimizing on a particular data type improves performance in any detector. SIGNIFICANCE: Automatic HFO detection is crucial to propel their clinical use as biomarkers of epileptogenic tissue. Comparing detectors on a single dataset is important to analyze their performance and to emphasize the issues involved in validation.
OBJECTIVE: High frequency oscillations (HFOs) are a biomarker of epileptogenicity. Visual marking of HFOs is highly time-consuming and inevitably subjective, making automatic detection necessary. We compare four existing detectors on the same dataset. METHODS: HFOs and baselines were identified by experienced reviewers in intracerebral EEGs from 20 patients. A new feature of our detector to deal with channels where baseline cannot be found is presented. The original and an optimal configuration are implemented. Receiver operator curves, false discovery rate, and channel ranking are used to evaluate performance. RESULTS: All detectors improve performance with the optimal configuration. Our detector had higher sensitivity, lower false positives than the others, and similar false detections. The main difference in performance was in very active channels. CONCLUSIONS: Each detector was developed for different recordings and with different aims. Our detector performed better in this dataset, but was developed on data similar to the test data. Moreover, optimizing on a particular data type improves performance in any detector. SIGNIFICANCE: Automatic HFO detection is crucial to propel their clinical use as biomarkers of epileptogenic tissue. Comparing detectors on a single dataset is important to analyze their performance and to emphasize the issues involved in validation.
Authors: Richard J Staba; Charles L Wilson; Anatol Bragin; Donald Jhung; Itzhak Fried; Jerome Engel Journal: Ann Neurol Date: 2004-07 Impact factor: 10.422
Authors: Richard J Staba; Leonardo Frighetto; Eric J Behnke; Gary W Mathern; Tony Fields; Anatol Bragin; Jennifer Ogren; Itzhak Fried; Charles L Wilson; Jerome Engel Journal: Epilepsia Date: 2007-07-28 Impact factor: 5.864
Authors: Birgit Frauscher; Fabrice Bartolomei; Katsuhiro Kobayashi; Jan Cimbalnik; Maryse A van 't Klooster; Stefan Rampp; Hiroshi Otsubo; Yvonne Höller; Joyce Y Wu; Eishi Asano; Jerome Engel; Philippe Kahane; Julia Jacobs; Jean Gotman Journal: Epilepsia Date: 2017-06-30 Impact factor: 5.864
Authors: Maryse A van 't Klooster; Frans S S Leijten; Geertjan Huiskamp; Hanneke E Ronner; Johannes C Baayen; Peter C van Rijen; Martinus J C Eijkemans; Kees P J Braun; Maeike Zijlmans Journal: Trials Date: 2015-09-23 Impact factor: 2.279
Authors: Catherine J Chu; Arthur Chan; Dan Song; Kevin J Staley; Steven M Stufflebeam; Mark A Kramer Journal: J Neurosci Methods Date: 2016-12-14 Impact factor: 2.390
Authors: Aljoscha Thomschewski; Nathalie Gerner; Patrick B Langthaler; Eugen Trinka; Arne C Bathke; Jürgen Fell; Yvonne Höller Journal: Front Neurol Date: 2020-10-19 Impact factor: 4.003