Smartwatch app spots 98% of seizures, cuts false alarms
A new study published in Neurology Open Access highlights the effectiveness of a smartwatch application in detecting tonic-clonic seizures among individuals with epilepsy. The research, led by Dr. James W. Wheless of Le Bonheur Children's Hospital in Memphis, Tennessee, found the EpiWatch app achieved a 98% detection rate while significantly reducing false alarms compared to existing devices. Tonic-clonic seizures, characterized by major convulsions, pose a serious risk to patients, particularly those who sleep alone. The condition can lead to loss of muscle tone and airway obstruction, contributing to Sudden Unexpected Death in Epilepsy, or SUDEP. Wearable devices are crucial for alerting caregivers to intervene, but high rates of false alarms have previously discouraged consistent use. This study aimed to determine if a common smartwatch with a dedicated app could overcome these challenges. The research involved 242 adults and children with epilepsy, with an average age of 23. Participants were monitored for an average of two and a half days within a specialized hospital unit equipped with video-electroencephalograph (EEG) technology. Neurologists classified seizure types based on this objective monitoring. All participants wore smartwatches running the EpiWatch app to track muscle activity. The results demonstrated high accuracy. Among 47 confirmed tonic-clonic seizures, the app detected 46, missing only one instance where a caregiver's physical restraint interfered with the sensor. Over 16,000 hours of monitoring, the system generated 56 false alarms, equating to 0.08 per day or one every 12.4 days. This rate is 90% lower than the 0.67 to 2.52 false alarms per day reported by other devices in the market. Additionally, 87% of participants experienced no false alarms at all. Most false positives were linked to repetitive movements, such as playing video games. Dr. Wheless noted that the overall sensitivity of 98% surpasses other devices, which typically range between 76% and 94%. A significant advantage of this approach is its appearance; unlike bulky medical wearables, a standard smartwatch does not carry the same social stigma, encouraging regular adoption. The app's reliability and discreet design are expected to promote long-term usage, providing caregivers with timely alerts to mitigate risks associated with severe seizures. The study found no adverse effects on the participants. However, researchers acknowledged a limitation: the controlled environment of an epilepsy monitoring unit may not fully reflect the variety of real-world conditions patients face. Despite this, the findings suggest that integrating seizure detection into everyday technology could significantly improve safety outcomes for people with epilepsy. The ability to provide reliable alerts with minimal interruption makes the EpiWatch app a promising tool in the management of tonic-clonic seizures.
