Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection
Wearable technology will become available and allow prolonged electroencephalography (EEG) monitoring in the home environment of patients with epilepsy. Neurologists analyse the EEG visually and annotate all seizures, which patients often under-report. Visual analysis of a 24-h EEG recording typical...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/4/1046 |
| _version_ | 1827603912527446016 |
|---|---|
| author | Thijs Becker Kaat Vandecasteele Christos Chatzichristos Wim Van Paesschen Dirk Valkenborg Sabine Van Huffel Maarten De Vos |
| author_facet | Thijs Becker Kaat Vandecasteele Christos Chatzichristos Wim Van Paesschen Dirk Valkenborg Sabine Van Huffel Maarten De Vos |
| author_sort | Thijs Becker |
| collection | DOAJ |
| description | Wearable technology will become available and allow prolonged electroencephalography (EEG) monitoring in the home environment of patients with epilepsy. Neurologists analyse the EEG visually and annotate all seizures, which patients often under-report. Visual analysis of a 24-h EEG recording typically takes one to two hours. Reliable automated seizure detection algorithms will be crucial to reduce this analysis. We investigated such algorithms on a dataset of behind-the-ear EEG measurements. Our first aim was to develop a methodology where part of the data is deferred to a human expert, who performs perfectly, with the goal of obtaining an (almost) perfect detection sensitivity (DS). Prediction confidences are determined by temperature scaling of the classification model outputs and trust scores. A DS of approximately 90% (99%) can be achieved when deferring around 10% (40%) of the data. Perfect DS can be achieved when deferring 50% of the data. Our second contribution demonstrates that a common modelling strategy, where predictions from several short EEG segments are combined to obtain a final prediction, can be improved by filtering out untrustworthy segments with low trust scores. The false detection rate shows a relative decrease between 21% and 43%, and the DS shows a small increase or decrease. |
| first_indexed | 2024-03-09T05:48:26Z |
| format | Article |
| id | doaj.art-b8a178f27ae84b6caf2dbb723806f6e2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T05:48:26Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-b8a178f27ae84b6caf2dbb723806f6e22023-12-03T12:19:28ZengMDPI AGSensors1424-82202021-02-01214104610.3390/s21041046Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure DetectionThijs Becker0Kaat Vandecasteele1Christos Chatzichristos2Wim Van Paesschen3Dirk Valkenborg4Sabine Van Huffel5Maarten De Vos6I-Biostat, Data Science Institute, Hasselt University, 3500 Hasselt, BelgiumSTADIUS Center for Dynamical Systems, Signal Processing and Data Analytics, Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, BelgiumSTADIUS Center for Dynamical Systems, Signal Processing and Data Analytics, Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, BelgiumDepartment of Neurology, UZ Leuven, 3001 Leuven, BelgiumI-Biostat, Data Science Institute, Hasselt University, 3500 Hasselt, BelgiumSTADIUS Center for Dynamical Systems, Signal Processing and Data Analytics, Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, BelgiumSTADIUS Center for Dynamical Systems, Signal Processing and Data Analytics, Department of Electrical Engineering (ESAT), KU Leuven, 3001 Leuven, BelgiumWearable technology will become available and allow prolonged electroencephalography (EEG) monitoring in the home environment of patients with epilepsy. Neurologists analyse the EEG visually and annotate all seizures, which patients often under-report. Visual analysis of a 24-h EEG recording typically takes one to two hours. Reliable automated seizure detection algorithms will be crucial to reduce this analysis. We investigated such algorithms on a dataset of behind-the-ear EEG measurements. Our first aim was to develop a methodology where part of the data is deferred to a human expert, who performs perfectly, with the goal of obtaining an (almost) perfect detection sensitivity (DS). Prediction confidences are determined by temperature scaling of the classification model outputs and trust scores. A DS of approximately 90% (99%) can be achieved when deferring around 10% (40%) of the data. Perfect DS can be achieved when deferring 50% of the data. Our second contribution demonstrates that a common modelling strategy, where predictions from several short EEG segments are combined to obtain a final prediction, can be improved by filtering out untrustworthy segments with low trust scores. The false detection rate shows a relative decrease between 21% and 43%, and the DS shows a small increase or decrease.https://www.mdpi.com/1424-8220/21/4/1046epilepsyseizure detectionelectroencephalographyclassification with a deferral optionhome monitoringlong-term monitoring |
| spellingShingle | Thijs Becker Kaat Vandecasteele Christos Chatzichristos Wim Van Paesschen Dirk Valkenborg Sabine Van Huffel Maarten De Vos Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection Sensors epilepsy seizure detection electroencephalography classification with a deferral option home monitoring long-term monitoring |
| title | Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection |
| title_full | Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection |
| title_fullStr | Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection |
| title_full_unstemmed | Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection |
| title_short | Classification with a Deferral Option and Low-Trust Filtering for Automated Seizure Detection |
| title_sort | classification with a deferral option and low trust filtering for automated seizure detection |
| topic | epilepsy seizure detection electroencephalography classification with a deferral option home monitoring long-term monitoring |
| url | https://www.mdpi.com/1424-8220/21/4/1046 |
| work_keys_str_mv | AT thijsbecker classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT kaatvandecasteele classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT christoschatzichristos classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT wimvanpaesschen classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT dirkvalkenborg classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT sabinevanhuffel classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection AT maartendevos classificationwithadeferraloptionandlowtrustfilteringforautomatedseizuredetection |