The development of mobile health and wearable technologies has recently boosted the field of seizure detection, in particular that of generalized tonic–clonic seizures (GTCS) using non–electroencephalography (EEG)-based methods.1-3 Growing clinical interest in developing these methods stems from their potential to trigger alarms, prompt timely interventions, and possibly reduce the risk of sudden unexpected death in epilepsy (SUDEP) issues tackled in this supplement.4, 5 By offering more objective seizure counting, seizure-detection devices might also allow for better evaluation of the effectiveness of antiepileptic treatment, both in clinical practice and studies. Such long-term companions could also provide reliable biomarkers for various comorbidities, including the risk of SUDEP, as well as novel insights into the dynamics of seizure recurrence in relation to various environmental and internal factors. This field has a number of limitations and challenges, including the paucity of high-quality studies and lack of well-established standards; the restricted seizure types that can be currently detected with appropriate sensitivity; the usability and acceptability of the available devices, including their rates of false alarms and battery recharging; and the need for multimodal sensing with more complex algorithms. The first international meeting on seizure detection and mobile health devices in epilepsy that took place in Copenhagen from July 6-8, 2017, was thus timely and successful in gathering more than 100 participants. This supplement of Epilepsia includes some of the most important data presented during the meeting. The primary methods used to detect seizures are detailed in reviews focusing on scalp EEG,6 surface electromyography (EMG),7 movement-based detection,8 and multimodal seizure detection.9 Scalp EEG offers the unique advantage of capturing most seizure types, with high sensitivity (75%-90%), but also a high rate of false alarms (0.1 and 5 per hour).6 It is not currently adapted to chronic ambulatory recordings, although this might change with the development of subcutaneously implanted electrodes. Surface EMG is typically used for detecting convulsive seizures, with 2 large-scale blinded prospective studies demonstrating high sensitivity (76%-100%) with average false-alarm rate ranging from 0.7 to 2.5/24 h.7 Movement-based GTCS detection, primarily using accelerometry sensors, is associated with highly variable sensitivity (31%-95%) and positive predictive value (4%-60%) across video-EEG studies, whereas a field study reported even lower sensitivity (14%).8 Conversely, 2 other field studies with high sensitivity are reported in this supplement: one using accelerometry showed a median sensitivity of 90% and a false alarm rate of only 0.1/d for GTCS detection,10 whereas another one using video-only in a residential care setting reported 100% sensitivity and a median false alarm rate of 0.78 per night.11 Multimodal seizure detection is currently characterized by various sensing methods and highly variable sensitivity (4%-100%) and rate of false alarms (0.25-20 per 8 hours).9 The large variability in the reported accuracy of the seizure-detection devices emphasizes the necessity to develop standards in designing and reporting clinical validation studies, an endeavor tackled in this supplement with a proposal of 5 study phases (0-4) based on 4 key features (subjects, recordings, data analysis, alarms, and reference standard).12 SB has served as a paid consultant for Brain Sentinel and UNEEG. PR has no conflicts of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.