Implementing Artificial-intelligence Wristbands to Help in Recording Seizures

The investigators hypothesize that the participants will be satisfied with artificial-intelligence wristband Embrace

Currently, seizure tracking relies on subjective patient and family recall. In several studies patients report only approximately half of their seizures, and even less during sleep. Sudden unexplained death in epilepsy (SUDEP) is the cause of premature death of up to 17% of all patients with epilepsy and as many as 50% with chronic refractory epilepsy (with rates as high as 1% per year). Its etiology is probably due to peri-ictal respiratory, cardiac, or autonomic nervous system dysfunction. Wristband sensors help caregivers assist patients during seizures and may reduce risks for complications such as injuries and SUDEP. Accurate seizure detection may improve the quality of life (QoL) of subjects and caregivers by decreasing burden of seizure monitoring and may facilitate diagnostic monitoring in the home setting. Possible risks are occurrence of alarm fatigue and invasion of privacy. The wristband Embrace has an overall sensitivity of 89.1% and an overall specificity of 93.1%. The investigators hypothesize that the participants will be satisfied with artificial-intelligence wristband Embrace. Methods: When the smart wristband detects the seizure, the mobile phone sends a text message to the companion and the nurse, the latter will see or get in touch with the companion and keep a record. In-patients receiving video-electroencephalography (VEEG) exam use Embrace during the VEEG recording which usually takes 3-5 days. Out-patients use Embrace for 8 weeks. The investigators will enroll 30-50 patients, half in-patients and half out-patients. The benefits are evaluated in terms of satisfaction. .

Inclusion Criteria: At least one seizure per month in the 3-month period before entering the study Exclusion Criteria: Uncountable seizures, psychogenic seizures, alcohol or drug abuse, severe heart or lung disorder, rapid progressive brain disorder, life-terminal disease, severe infection/ bleeding/ liver/ kidney disorder, platelet <80,000/μL, neutrophil <1,000/μL, clinically significant ECG abnormality, unstable psychiatric disease (except anxiety).