Tuesday, February 6, 2018

Smart watch for seizure tracking and epilepsy management.

The US Food and Drug Administration (FDA) has cleared for marketing the Embrace smart watch (Empatica Inc) for seizure tracking and epilepsy management.

The Embrace smart watch uses advanced machine learning to identify convulsive seizures and sends an alert via text and phone message to caregivers.

Embrace was tested in a clinical study involving 135 patients with epilepsy who were admitted to epilepsy monitoring units for continuous monitoring with video electroencephalography while simultaneously wearing the device, which records electrodermal activity.

Researchers collected 6530 hours' worth of data over 272 days, including 40 generalized tonic-clonic seizures. Embrace's algorithm detected 100% of the seizures, confirmed by independent epilepsy experts.  The device also records sleep, rest, and physical activity data. 

"The FDA approval of the Embrace device to detect major convulsive seizures represents a major milestone in the care of epilepsy patients," Orrin Devinsky, director of the Comprehensive Epilepsy Center at NYU Langone in New York City, said in a news release from the company.

"Tragically, more than 3000 Americans die each year from sudden unexpected death in epilepsy (SUDEP) and the Embrace offers the potential to alarm family members and caretakers that a tonic-clonic seizure is occurring. The scientific evidence strongly supports that prompt attention during or shortly after these convulsive seizures can be life-saving in many cases," Dr Devinsky added.

Embrace was approved in Europe for seizure monitoring and alerts in April 2017.

The US Centers for Disease Control and Prevention estimates that about 3.4 million people in the United States have epilepsy, including 470,000 children.


Onorati F, Regalia G, Caborni C, Migliorini M, Bender D, Poh MZ, Frazier C, Kovitch Thropp E, Mynatt ED, Bidwell J, Mai R, LaFrance WC Jr, Blum AS, Friedman D, Loddenkemper T, Mohammadpour-Touserkani F, Reinsberger C, Tognetti S, Picard RW. Multicenter clinical assessment of improved wearable multimodal convulsive seizure detectors. Epilepsia. 2017 Nov;58(11):1870-1879.


New devices are needed for monitoring seizures, especially those associated with sudden unexpected death in epilepsy (SUDEP). They must be unobtrusive and automated, and provide false alarm rates (FARs) bearable in everyday life. This study quantifies the performance of new multimodal wrist-worn convulsive seizure detectors.

Hand-annotated video-electroencephalographic seizure events were collected from 69 patients at six clinical sites. Three different wristbands were used to record electrodermal activity (EDA) and accelerometer (ACM) signals, obtaining 5,928 h of data, including 55 convulsive epileptic seizures (six focal tonic-clonic seizures and 49 focal to bilateral tonic-clonic seizures) from 22 patients. Recordings were analyzed offline to train and test two new machine learning classifiers and a published classifier based on EDA and ACM. Moreover, wristband data were analyzed to estimate seizure-motion duration and autonomic responses.

The two novel classifiers consistently outperformed the previous detector. The most efficient (Classifier III) yielded sensitivity of 94.55%, and an FAR of 0.2 events/day. No nocturnal seizures were missed. Most patients had <1 false alarm every 4 days, with an FAR below their seizure frequency. When increasing the sensitivity to 100% (no missed seizures), the FAR is up to 13 times lower than with the previous detector. Furthermore, all detections occurred before the seizure ended, providing reasonable latency (median = 29.3 s, range = 14.8-151 s). Automatically estimated seizure durations were correlated with true durations, enabling reliable annotations. Finally, EDA measurements confirmed the presence of postictal autonomic dysfunction, exhibiting a significant rise in 73% of the convulsive seizures.

The proposed multimodal wrist-worn convulsive seizure detectors provide seizure counts that are more accurate than previous automated detectors and typical patient self-reports, while maintaining a tolerable FAR for ambulatory monitoring. Furthermore, the multimodal system provides an objective description of motor behavior and autonomic dysfunction, aimed at enriching seizure characterization, with potential utility for SUDEP warning.


  1. Picard RW, Migliorini M, Caborni C, Onorati F, Regalia G, Friedman D, Devinsky O. Wrist sensor reveals sympathetic hyperactivity and hypoventilation before probable SUDEP. Neurology. 2017 Aug 8;89(6):633-635.

    We report a probable sudden unexpected death in epilepsy (SUDEP) in a 20-year-old man wearing a smartwatch that recorded wrist motion via 3-axis accelerometer (ACC) and electrodermal activity (EDA). EDA reflects sympathetic activity without parasympathetic antagonism.1 The smartwatch (Empatica [Milan, Italy] Embrace, with CE Medical clearance from the European Union for seizure detection) issued an alert, received by the caregiver at 8:50 AM, indicating a probable convulsive seizure. An adult trained in cardiopulmonary resuscitation (CPR) arrived at 9:05 AM, found the patient pulseless, prone, face in pillow with mucus in his mouth, and commenced CPR for 15 minutes without recovery. The family declined autopsy.

    The patient had been diagnosed at age 5 with epilepsy. He had 3–4 generalized tonic-clonic seizures (GTCS) a year, typically during mornings before awakening. More than 6 antiepileptic medications failed to control seizures. Routine and prolonged EEGs were normal. At the time of death, he was taking lamotrigine twice daily and extended-release oxcarbazepine each night…

    This case obtained continuous EDA 24/7, finding an unusually large (66.7 µS) EDR [postictal electrodermal response] preceding death in a probable case of SUDEP. EDRs during nonterminal GTCS in drug-resistant patients (ages 3–20) showed peaks from 3.9 to 24.3 µS,3 which are within ranges occurring with vigorous sports or natural stressors. While this EDR was highly elevated, it is unknown whether EDR changes of this magnitude are a specific SUDEP signature or could occur with other forms of sudden death.

    EDRs can be elicited by electrical stimulation of the amygdala, hippocampus, and cingulate gyrus.4 The physiologic basis for postictal EDR elevation is not fully understood but may reflect hypothalamic or brainstem centers released from cortical inhibition.(continued)

  2. (continued) The extremely elevated peri-ictal EDR at the time of death in this case supports autonomic dysfunction and severe postictal cerebral dysfunction as SUDEP mechanisms. Sympathetic hyperactivity following GTCS and parasympathetic hyperactivity in animal models of SUDEP suggest that pathologic coactivation of opposing systems may contribute to SUDEP. Prolonged postictal generalized EEG suppression (PGES; scalp EEG channels <10 µV), a putative biomarker for SUDEP risk, occurred in 11/11 EEG-monitored SUDEPs immediately after a GTCS and prolonged generalized EEG suppression was observed before 3/3 SUDEPs without evidence of a seizure. In non-SUDEPs, EDR magnitude accompanying a GTCS correlates with PGES duration; furthermore, all GTCS with EDR amplitude greater >=15 µS exhibited prolonged PGES >20 seconds. Thus, finding an unusually large EDR with this terminal GTCS is consistent with profound sympathetic hyperactivity and prolonged PGES before death.

    These data reveal a critical window of postictal autonomic dysregulation, probably relevant to SUDEP pathophysiology, that is detectable with a wristband sensor. While medical support, even within 2 minutes of a seizure, does not always prevent SUDEP, evidence supports that SUDEP is less likely to occur when someone can rapidly stimulate or reposition. Frequent GTCS and long duration of epilepsy placed this patient at very high risk for SUDEP. Despite having seen 6 neurologists, the family was never told about the possibility of SUDEP; they expected the patient to recover from all of his seizures. Patients and families should be informed about SUDEP risk and strategies to reduce risk (e.g., medication adherence, sleep hygiene). They should be told that SUDEP can happen in the minutes following a seizure, and if somebody is there to immediately stimulate, reposition, or provide first aid, the patient may have a better chance of survival. For high-risk patients, seizure detection and alert monitors may bring help. As this case illustrates, their utility in preventing SUDEP is predicated on a helper arriving quickly and providing appropriate aid.

  3. Beniczky S, Conradsen I, Henning O, Fabricius M, Wolf P. Automated real-time detection of tonic-clonic seizures using a wearable EMG device. Neurology. 2018 Jan 30;90(5):e428-e434.


    To determine the accuracy of automated detection of generalized tonic-clonic seizures (GTCS) using a wearable surface EMG device.

    We prospectively tested the technical performance and diagnostic accuracy of real-time seizure detection using a wearable surface EMG device. The seizure detection algorithm and the cutoff values were prespecified. A total of 71 patients, referred to long-term video-EEG monitoring, on suspicion of GTCS, were recruited in 3 centers. Seizure detection was real-time and fully automated. The reference standard was the evaluation of video-EEG recordings by trained experts, who were blinded to data from the device. Reading the seizure logs from the device was done blinded to all other data.

    The mean recording time per patient was 53.18 hours. Total recording time was 3735.5 hours, and device deficiency time was 193 hours (4.9% of the total time the device was turned on). No adverse events occurred. The sensitivity of the wearable device was 93.8% (30 out of 32 GTCS were detected). Median seizure detection latency was 9 seconds (range -4 to 48 seconds). False alarm rate was 0.67/d.

    The performance of the wearable EMG device fulfilled the requirements of patients: it detected GTCS with a sensitivity exceeding 90% and detection latency within 30 seconds.

    This study provides Class II evidence that for people with a history of GTCS, a wearable EMG device accurately detects GTCS (sensitivity 93.8%, false alarm rate 0.67/d).