An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections
Abstract Objective Aiming to improve the feasibility and reliability of using high‐frequency oscillations (HFOs) for translational studies of epilepsy, we present a pipeline with features specifically designed to reject false positives for HFOs to improve the automatic HFO detector. Methods We prese...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-12-01
|
| Series: | Epilepsia Open |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/epi4.12647 |
| _version_ | 1811321757801381888 |
|---|---|
| author | Yufeng Zhou Jing You Udaya Kumar Shennan A Weiss Anatol Bragin Jerome Engel Jr Christos Papadelis Lin Li |
| author_facet | Yufeng Zhou Jing You Udaya Kumar Shennan A Weiss Anatol Bragin Jerome Engel Jr Christos Papadelis Lin Li |
| author_sort | Yufeng Zhou |
| collection | DOAJ |
| description | Abstract Objective Aiming to improve the feasibility and reliability of using high‐frequency oscillations (HFOs) for translational studies of epilepsy, we present a pipeline with features specifically designed to reject false positives for HFOs to improve the automatic HFO detector. Methods We presented an integrated, multi‐layered procedure capable of automatically rejecting HFOs from a variety of common false positives, such as motion, background signals, and sharp transients. This method utilizes a time‐frequency contour approach that embeds three different layers including peak constraints, power thresholds, and morphological identification to discard false positives. Four experts were involved in rating detected HFO events that were randomly selected from different posttraumatic epilepsy (PTE) animals for a comprehensive evaluation. Results The algorithm was run on 768‐h recordings of intracranial electrodes in 48 PTE animals. A total of 453 917 HFOs were identified by initial HFO detection, of which 450 917 were implemented for HFO refinement and 203 531 events were retained. Random sampling was used to evaluate the performance of the detector. The HFO detection yielded an overall accuracy of 0.95±0.03, with precision, recall, and F1 scores of 0.92±0.05, 0.99±0.01, and 0.94±0.03, respectively. For the HFO classification, our algorithm obtained an accuracy of 0.97±0.02. For the inter‐rater reliability of algorithm evaluation, the agreement among four experts was 0.94±0.03 for HFO detection and 0.85±0.04 for HFO classification. Significance Our approach shows that a segregated pipeline design with a focus on false‐positive rejection can improve the detection efficiency and provide reliable results. This pipeline does not require customization and uses fixed parameters, making it highly feasible and translatable for basic and clinical applications of epilepsy. |
| first_indexed | 2024-04-13T13:23:18Z |
| format | Article |
| id | doaj.art-e2f6dbc0a32d4f198545f799248f5031 |
| institution | Directory Open Access Journal |
| issn | 2470-9239 |
| language | English |
| last_indexed | 2024-04-13T13:23:18Z |
| publishDate | 2022-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | Epilepsia Open |
| spelling | doaj.art-e2f6dbc0a32d4f198545f799248f50312022-12-22T02:45:16ZengWileyEpilepsia Open2470-92392022-12-017467468610.1002/epi4.12647An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detectionsYufeng Zhou0Jing You1Udaya Kumar2Shennan A Weiss3Anatol Bragin4Jerome Engel Jr5Christos Papadelis6Lin Li7Department of Biomedical Engineering University of North Texas Texas USADepartment of Biomedical Engineering University of North Texas Texas USADepartment of Neurology University of California Los Angeles Los Angeles California USADepartments of Neurology, Department of Physiology and Pharmacology State University of New York Downstate Brooklyn New York USADepartment of Neurology University of California Los Angeles Los Angeles California USADepartment of Neurology University of California Los Angeles Los Angeles California USAJane and John Justin Neurosciences Center Cook Children's Health Care System Fort Worth Texas USADepartment of Biomedical Engineering University of North Texas Texas USAAbstract Objective Aiming to improve the feasibility and reliability of using high‐frequency oscillations (HFOs) for translational studies of epilepsy, we present a pipeline with features specifically designed to reject false positives for HFOs to improve the automatic HFO detector. Methods We presented an integrated, multi‐layered procedure capable of automatically rejecting HFOs from a variety of common false positives, such as motion, background signals, and sharp transients. This method utilizes a time‐frequency contour approach that embeds three different layers including peak constraints, power thresholds, and morphological identification to discard false positives. Four experts were involved in rating detected HFO events that were randomly selected from different posttraumatic epilepsy (PTE) animals for a comprehensive evaluation. Results The algorithm was run on 768‐h recordings of intracranial electrodes in 48 PTE animals. A total of 453 917 HFOs were identified by initial HFO detection, of which 450 917 were implemented for HFO refinement and 203 531 events were retained. Random sampling was used to evaluate the performance of the detector. The HFO detection yielded an overall accuracy of 0.95±0.03, with precision, recall, and F1 scores of 0.92±0.05, 0.99±0.01, and 0.94±0.03, respectively. For the HFO classification, our algorithm obtained an accuracy of 0.97±0.02. For the inter‐rater reliability of algorithm evaluation, the agreement among four experts was 0.94±0.03 for HFO detection and 0.85±0.04 for HFO classification. Significance Our approach shows that a segregated pipeline design with a focus on false‐positive rejection can improve the detection efficiency and provide reliable results. This pipeline does not require customization and uses fixed parameters, making it highly feasible and translatable for basic and clinical applications of epilepsy.https://doi.org/10.1002/epi4.12647complex waveletepilepsyhigh‐frequency oscillationstopographical analysis |
| spellingShingle | Yufeng Zhou Jing You Udaya Kumar Shennan A Weiss Anatol Bragin Jerome Engel Jr Christos Papadelis Lin Li An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections Epilepsia Open complex wavelet epilepsy high‐frequency oscillations topographical analysis |
| title | An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections |
| title_full | An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections |
| title_fullStr | An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections |
| title_full_unstemmed | An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections |
| title_short | An approach for reliably identifying high‐frequency oscillations and reducing false‐positive detections |
| title_sort | approach for reliably identifying high frequency oscillations and reducing false positive detections |
| topic | complex wavelet epilepsy high‐frequency oscillations topographical analysis |
| url | https://doi.org/10.1002/epi4.12647 |
| work_keys_str_mv | AT yufengzhou anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT jingyou anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT udayakumar anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT shennanaweiss anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT anatolbragin anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT jeromeengeljr anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT christospapadelis anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT linli anapproachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT yufengzhou approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT jingyou approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT udayakumar approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT shennanaweiss approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT anatolbragin approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT jeromeengeljr approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT christospapadelis approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections AT linli approachforreliablyidentifyinghighfrequencyoscillationsandreducingfalsepositivedetections |