Semiautomated classification of nocturnal seizures using video recordings

Abstract

ObjectiveThe objective of this study was to evaluate the accuracy of a semiautomated classification of nocturnal seizures using a hybrid system consisting of an artificial intelligence‐based algorithm, which selects epochs with potential clinical relevance to be reviewed by human experts.MethodsConsecutive patients with nocturnal motor seizures admitted for video‐electroencephalographic long‐term monitoring (LTM) were prospectively recruited. We determined the extent of data reduction by using the algorithm, and we evaluated the accuracy of seizure classification from the hybrid system compared with the gold standard of LTM.ResultsForty consecutive patients (24 male; median age = 15 years) were analyzed. The algorithm reduced the duration of epochs to be reviewed to 14% of the total recording time (1874 h). There was a fair agreement beyond chance in seizure classification between the hybrid system and the gold standard (agreement coefficient = .33, 95% confidence interval = .20–.47). The hybrid system correctly identified all tonic–clonic and clonic seizures and 82% of focal motor seizures. However, there was low accuracy in identifying seizure types with more discrete or subtle motor phenomena.SignificanceUsing a hybrid (algorithm–human) system for reviewing nocturnal video recordings significantly decreased the workload and provided accurate classification of major motor seizures (tonic–clonic, clonic, and focal motor seizures).

Clinical utility of a video/audio-based epilepsy monitoring system Nelli

“…Since the Nelli-system can register only seizures with alterations in motor activity, patients with nonmotor seizures cannot be evaluated. In the validation study of the Nelli Ò hybrid system, there was lower accuracy in classifying more discrete motor seizures such as myoclonic jerks, short tonic seizures, and epileptic spasms [18]. On the other hand, the validation study included also children and adolescents [18], but the present retrospective study only included patients aged 16 or older, in which these seizure types are less common.…”

“…In the validation study of the Nelli Ò hybrid system, there was lower accuracy in classifying more discrete motor seizures such as myoclonic jerks, short tonic seizures, and epileptic spasms [18]. On the other hand, the validation study included also children and adolescents [18], but the present retrospective study only included patients aged 16 or older, in which these seizure types are less common. Although multiple registrations were performed for some patients, we only reported the outcome from the first registration.…”

“…According to our recent validation study the Nelli Ò hybrid system was able to provide accurate classification of motor seizures including tonic-clonic, clonic, and focal motor seizures [18]. The semiology of these seizures could also be further characterized to include descriptions of automatisms, clonic, hyperkinetic, myoclonic, and tonic activity.…”

“…The Nelli Ò seizure monitoring system is an audio/video-based semi-automatic (hybrid) seizure monitoring platform that uses computer vision and machine learning to identify kinematic data commonly associated with seizures with a positive motor component and human experts to visually assess these epochs [16,17]. In a recent validation study, the Nelli Ò hybrid system was used in a blinded setting without any prior information on the patients or their seizure types against video-EEG monitoring at a well-established epilepsy center providing accurate classification of major motor seizures including tonic-clonic, clonic, and focal motor seizures [18]. CE-marked Nelli Ò hybrid system has been recommended for clinical use in Finland by a government set body (the National Coordinating Group for Drug-resistant Epilepsy).…”

The unforeseen future: Impacts of the COVID‐19 pandemic on home video‐EEG telemetry

“…Potentially more readily available in a wide variety of diagnostic and monitoring settings, remote video monitoring with human analysis (such as the vCreate platform adopted at Evelina Children's Hospital, Section 3.3.2 ) or partially automated video analysis with dedicated hardware (such as the system offered by NeuroEvent Labs 31 , 32 ).…”