Alba Sierra-Marcos1, Mark L Scheuer2, Andrea O Rossetti3. 1. Department of Clinical Neurosciences, Centre Hospitalier Universitaire Vaudois (CHUV), and University of Lausanne, Lausanne, Switzerland. 2. Persyst Development Corporation, San Diego, CA, USA. 3. Department of Clinical Neurosciences, Centre Hospitalier Universitaire Vaudois (CHUV), and University of Lausanne, Lausanne, Switzerland. Electronic address: andrea.rossetti@chuv.ch.
Abstract
OBJECTIVE: To evaluate an automated seizure detection (ASD) algorithm in EEGs with periodic and other challenging patterns. METHODS: Selected EEGs recorded in patients over 1year old were classified into four groups: A. Periodic lateralized epileptiform discharges (PLEDs) with intermixed electrical seizures. B. PLEDs without seizures. C. Electrical seizures and no PLEDs. D. No PLEDs or seizures. Recordings were analyzed by the Persyst P12 software, and compared to the raw EEG, interpreted by two experienced neurophysiologists; Positive percent agreement (PPA) and false-positive rates/hour (FPR) were calculated. RESULTS: We assessed 98 recordings (Group A=21 patients; B=29, C=17, D=31). Total duration was 82.7h (median: 1h); containing 268 seizures. The software detected 204 (=76.1%) seizures; all ictal events were captured in 29/38 (76.3%) patients; in only in 3 (7.7%) no seizures were detected. Median PPA was 100% (range 0-100; interquartile range 50-100), and the median FPR 0/h (range 0-75.8; interquartile range 0-4.5); however, lower performances were seen in the groups containing periodic discharges. CONCLUSION: This analysis provides data regarding the yield of the ASD in a particularly difficult subset of EEG recordings, showing that periodic discharges may bias the results. SIGNIFICANCE: Ongoing refinements in this technique might enhance its utility and lead to a more extensive application.
OBJECTIVE: To evaluate an automated seizure detection (ASD) algorithm in EEGs with periodic and other challenging patterns. METHODS: Selected EEGs recorded in patients over 1year old were classified into four groups: A. Periodic lateralized epileptiform discharges (PLEDs) with intermixed electrical seizures. B. PLEDs without seizures. C. Electrical seizures and no PLEDs. D. No PLEDs or seizures. Recordings were analyzed by the Persyst P12 software, and compared to the raw EEG, interpreted by two experienced neurophysiologists; Positive percent agreement (PPA) and false-positive rates/hour (FPR) were calculated. RESULTS: We assessed 98 recordings (Group A=21 patients; B=29, C=17, D=31). Total duration was 82.7h (median: 1h); containing 268 seizures. The software detected 204 (=76.1%) seizures; all ictal events were captured in 29/38 (76.3%) patients; in only in 3 (7.7%) no seizures were detected. Median PPA was 100% (range 0-100; interquartile range 50-100), and the median FPR 0/h (range 0-75.8; interquartile range 0-4.5); however, lower performances were seen in the groups containing periodic discharges. CONCLUSION: This analysis provides data regarding the yield of the ASD in a particularly difficult subset of EEG recordings, showing that periodic discharges may bias the results. SIGNIFICANCE: Ongoing refinements in this technique might enhance its utility and lead to a more extensive application.
Authors: Dominique Duncan; Paul Vespa; Asla Pitkänen; Adebayo Braimah; Niina Lapinlampi; Arthur W Toga Journal: Neurobiol Dis Date: 2018-06-01 Impact factor: 5.996
Authors: J G Bogaarts; D M W Hilkman; E D Gommer; V H J M van Kranen-Mastenbroek; J P H Reulen Journal: Med Biol Eng Comput Date: 2016-04-06 Impact factor: 2.602
Authors: Erica D McKenzie; Andrew S P Lim; Edward C W Leung; Andrew J Cole; Alice D Lam; Ani Eloyan; Damber K Nirola; Lhab Tshering; Ronald Thibert; Rodrigo Zepeda Garcia; Esther Bui; Sonam Deki; Liesly Lee; Sarah J Clark; Joseph M Cohen; Jo Mantia; Kate T Brizzi; Tali R Sorets; Sarah Wahlster; Mia Borzello; Arkadiusz Stopczynski; Sydney S Cash; Farrah J Mateen Journal: Sci Rep Date: 2017-04-03 Impact factor: 4.379