Table 2.

Categories of non-EEG based approaches for seizure detection, seizure prediction, and seizure forecasting (illustrated in Figure 7).

Non-EEG approaches	Target use case	Duration of use	Sensitivity	False positive rate	Key challenges
Bed mattress	Nocturnal convulsive seizures and SUDEP	Months to Years	62–89%	3/year to 0.5/night	Nonconvulsive seizures, sleep behaviors, sensor placement
Arm-worn (biceps)	Convulsive seizures	Days to Months	75–90%	0.5/day to 6/night	Wearability, correct sensor placement, movement artifacts
Wristwatch	Convulsive seizures	Days to Months	80–95%	0.25/day to 1.2/day	Wearability, movement artifacts, noisy data
In Ear*	Convulsive and Electrographic seizures	Hours to Days	55–99%	5–60% of all detections	Wearability for long term use, restricted to near-ear epilepsy

See text for citations of specific technologies. The in-ear technology can use EEG in addition to non-EEG signals for seizure detection.