This MIT Startup created the first FDA cleared smartwatch for epilepsy patients
The MIT startup Empatica, developed the first FDA cleared multimodal smartwatch for epilepsy patients, was approved as a medical device in 2018. The product was named ‘Embrace’. It’s a seizure-alerting smartwatch sensing the physiological signals of ongoing Generalized Tonic Clonic seizure(GTCS-like events), then send alert to caregivers, which can be a life-saving device for patients who are under high risk of sudden unexpected death in epilepsy(SUDEP), in Generalized Tonic Clonic Seizure (GTCS). Traditionally GTCS can only be detected by video with EEG in hospitals and clinics. The company made this breakthrough by making seizure detection viable in daily life at the convenience of wearing a stylish smartwatch.
The smartwatch Embrace was equipped with accelerometer and EDA sensor, with machine learning algorithm, to detect convulsive seizure-like motion and EDA signals on the skin.
It is Rosalind Picard, founder of the company Empatica and director of Affective Computing Research Lab at MIT, who pioneers the field of research on the use of Electrodermal activity (EDA signals) on the skin for seizure detection (specifically GTCS). Rosalind Picard found out that during onset of a seizure, the patient’s EDA signals goes extraordinary high resulting from arousal of sympathetic nervous system. This results in changes in the state of sweat glands and changes in electrical conductance on the skin, causing the EDA signals to go high. Moreover, EDA signal is able to detect brain activity involved in seizure and anxiety which sometimes even EEG cannot detect if the seizure involves areas that are deep inside the brain. In times of seizure when patients undergo emotional stress and physical exertion, the brain will send signals to the skin causing changes in electrical conductance (electrodermal activity) across the skin. The smartwatch detects a high EDA signals coupled with convulsive seizure-like motion, the
The product shows promising results in multimodal sensing algorithm compared to unimodal in seizure detection. There have been other products in the market that uses only motion detection (accelerometry alone) for seizure detection, but has lower sensitivity then this Embrace smartwatch. According to the research funded by Dutch Epilepsy Fund/National Science Foundation. sensitivity of seizure detection is only 88% if using acceleromtry alone, compared to 94% sensitivity which uses both accelerometry and electrodermal activity, This proves that multimodal may work better than unimodal sensing.
Whilst EDA signals and motion detection has a higher sensitivity but it’s not foolproof. There are both false positives and negatives. According to a few user review,
it causes problems if the person has a very active lifestyle since the smartwatch is incredibly sensitive to motion. There are false alarms when the wearers are having intense physical activity such as running, and doing exercise. These users recommended that the product may only be suitable for someone having a less physically active lifestyle, as well as having multiple, higher debilitating seizure a day. It makes sense that Empatica makes the smartwatch as a prescription only device as it’s best to consult the doctor before deciding whether it’s a right device for the epilepsy patient or not.
In fact, some activities and bodily movements in daily life may seem like convulsive seizure to the smartwatch, resulting in false alarms. According to Empatica, the motion of convulsive seizure takes precedence in Embrace smartwatch’s algorithm. EDA also takes part in the smartwatch’s algorithm, but since EDA is more prone to fluctuate, relying too heavily on EDA signals would result in even more false alarms. It’s good that in latest review article in the product it shows there has been continuous improvement on the algorithm owing to growing data availability and more users contributed immediate feedback on the false alarm problem, thereby facilitating machine learning algorithm adjustment to better distinguish normal physical activities from GTCS. The false alarm rate has been lowered from the initial ~2 to 0.2–1 false alarms per day.
Apart from false positive alarm, it has false negative as well. One user reports the Embrace smartwatch was unable to alert caregivers when the patient was having grand mal seizures. It is possible that during the tonic phase of the generalized tonic-clonic seizure when the muscle clenches, movement may be so subtle that the accelerometer of the smartwatch or other motion detectors wouldn’t be able to pick up the occurring seizure. Another company Brain Sentinel has done research in this area. According to the company, during the tonic phase of the GTCS, surface Electromyography(sEMG) is a better alternative to other motion detector like accelerometer for seizure detection. It is because sEMG manages to detect vigorous muscle activity unique to the tonic phase of GTCS. It doesn’t rely on any other physiological measurements such as heart rate variability, EDA…etc.
Brain Sentinel has developed a device called SPEAR System which has received FDA clearance for GTCS detection based on sEMG. It is a device to be worn on the biceps. During onset of a seizure there would be muscle signals on the biceps that comes from the part of the brain which causes the seizure event. When the muscles is activated during seizure, the device senses the muscle clenches and tightens which then send alarm to caregivers. Many types of seizures have motor manifestations, which has led to the analysis of muscle activity and movements for seizure detection.
As for the accuracy of the device, the Epilepsy Foundation revealed that data from 29 people with generalized tonic-clonic seizures identified by three independent epilepsy specialists revealed the SPEAR System had 100% sensitivity for detecting seizures. This however doesn’t represent the whole user population. One study shows sEMG device today is not totally foolproof. In a clinical trial conducted by Department of Neurophysiology in S.B. Danish Epilepsy Center with 71 patients wearing sEMG device, the results shows a sensitivity of 93.8% and false alarm rate of 0.67 alarm/day.
The above reveals that false-positive and false-negative is a challenge for seizure detection device for daily use. One reason to explain this is that seizures events may be different for each individual and seizure patterns fluctuate because of external environment. There need to be more data collected from wide variety of real life situation from users in order to facilitate continuous improvement of the algorithm.
As we can see both products above (SPEAR System and Embrace smartwatch), using different physiological sensors. This has led me to thinking about combining all the physiological measurement together in one device so we could profit from the strengths of each of the measurement. Wouldn’t it be possible?
According to the research funded by the Dutch Epilepsy Fund/National Science Foundation which examine combined use of multiple physiological sensors for seizure detection in a few studies (accelerometers, electromyography(EMG), heart rate, and oximetry, Electrodermal activity(EDA)..etc.) In these studies, It was found that multiple sensors increased sensitivity, with false detections decreased in a study, but increased in another study. Why is it so?
In fact, while combining different sensors may improve detection accuracy and lower false alarms in some circumstances, it is not always the case. Often when a device becomes too sensitive, the number of false-positive detections(false alarm) may rise. It is also possible with too many sensors, signals can interfere with each other.
The study has shown promising sensitivity for multimodal devices, but minimizing the number of false positives is still challenging. According to Frans S.S. Leijten of the Dutch TeleEpilepsy Consortium , false alarm is still a major problem in the devices and is the major impediment for clinical devices to be used at home.
What is encouraging is that the false detections seem to occur only in a minority of patients (around 16%-30% of patients).
In fact, the false detections occurring only in minority of patients opens up the prospect of using generic algorithm in a device for most patients, and personalization of the algorithm for the few patients who turns out to have many false detections. To market these devices in the consumer market, this should be part of the instructions manual and user information before purchase.
Another drawback for multimodal seizure detection device is that the device can become too bulky and heavy if there are too many sensors. Proper positioning of the sensors is important for accuracy. In nowadays’ technology, different sensors may need to be placed in different parts of the body for various physiological measurements, which is not convenient to be used daily. I anticipate a breakthrough in future when we can combine all these physiological measurements (sEMG, Heart rate variability ECG, EDA signal, oximetry, EEG….etc) in one single light-weighted device, a easy-to-use in-home seizure detection device that would improve the quality of life of patients and caregivers.
Another limitation of these seizure detection devices today is that it needs to pair with a smartphone within certain distances of the wearer’s device in order to facilitate alert messages being sent to caregivers via the mobile phone. The device cannot send the alert message without the mobile phone. This poses inconvenience to wearers if they run around and leave their mobile phone somewhere. In fact, development of these products is limited by today’s smartwatch technology: Even Apple Watch have limited functionality if used alone without an iphone. There would need to have another breakthrough on the hardware before a smartwatch like Empatica’s Embrace can do more with the size of a small size wristwatch.
With Empatica’s Embrace smartwatch, it can monitor sleep and stress patterns apart from detecting seizure. But some epilepsy patients may expect a smartwatch to do more, acting as a mini caretaker, such as monitoring their nervous system, or even warn them of an impending seizure. Today, this is still limited by the lack of biomarkers that can predict/diagnose Epilepsy. The patient won’t know he/she is a Epilepsy patient until he/she undergoes the cumbersome process of a video-EEG(Electroencephalography) in the hospital/clinic after he/she has the first and second seizure. The Epilepsy Foundation is still looking for potential biomarkers which can diagnose a epileptic seizure before the patient has the first one. I hope there will be a breakthrough in this aspect which would be good news to the patients. I look forward to seeing the next generation of seizure detection device(probably in next decade), with breakthrough on both the hardware (light weighted multimodal sensing in one single piece) and a biomarker to notify the patient before the seizure takes place.
One thing I appreciate is that Empatica treats the physiological and personal data they collected with respect and would never use it for any research without consent of their customers. The company is fully compliant with GDPR data protection and has extended it to customers worldwide.