More Than a Watch: How Wearable Tech is Helping Advance Neuroscience

Post by Lincoln Tracy

What's the science?

Wearable technology is a fast-moving and financially lucrative industry. The adoption of this technology is primarily being driven by smartwatches. These devices can measure a range of physiological signals including heart rate, blood oxygen levels, and sweat gland activation – all while looking sleek and stylish on your wrist. Smartwatches allow scientists to collect a wide range of long-term, real-time physiological data from large numbers of people at the same time. This week in Neuron, Johnson and Picard provide a broad overview of wearable technology and highlight its potential clinical applications.

What have we learned?

One type of smartwatch data is electrodermal activity, or EDA. EDA refers to changes in the electrical properties of our skin. When our sympathetic nervous system is activated, we begin to sweat. These increases in sympathetic activity can be measured by passing a small electric current across two electrodes on the surface of the skin and calculating the electrical conductance, giving us a measure of EDA. Measuring EDA allows scientists to examine purely sympathetic activity, as EDA does not have any known parasympathetic drivers.

However, EDA is more than a simple measure of arousal. One wider-ranging application of EDA is the detection and characterization of seizures, particularly generalized tonic-clonic seizures. The association between EDA and seizure activity has been identified from concurrent electroencephalography (EEG) and EDA recordings from patients having a seizure. The recordings show unusual electrical activity across all EEG channels during the seizure. But after the seizure stops, there is a flattening of EEG activity and a surge in EDA. Several studies in both children and adults have shown that measures of EDA correlate with the duration of the post-seizure suppression of EEG activity.

Why does it matter?

Correlations such as these allow for major advances in real-world diagnostic tests and interventions. For example, EEG suppression has been observed for all EEG-monitored cases of sudden unexpected death in epilepsy (SUDEP), the second leading cause of years of potential lives lost for neurological conditions. The exact mechanisms of SUDEP are unknown, but the risk of SUDEP is greatest with frequent seizures and being alone at the time of seizure. However, by applying machine learning to EDA and accelerometry data, life-threatening seizures can be detected in real-time with almost 100% sensitivity. Detecting the seizures before they end allows caregivers and loved ones to attend to the person seizing and provide aid, potentially preventing death.

What are the next steps?

Despite EDA being studied since the 19th century, the advent and popularity of wearable technology allow scientists to explore the role of this (and other) physiological signals in everyday life. The simplicity and practicality of smartwatches allow their use in vulnerable or underserved populations where typical neuroimaging technologies cannot be used. But it is important to note that wearable technology is not without its limitations. Many devices have their own proprietary algorithms, meaning that the underlying raw data is kept under strict lock and key. There are also issues surrounding the privacy, transparency, and ethical use of the data collected by these devices. These limitations aside, the addition of wearable technology to research kits and combining their use with existing neuroimaging techniques offer the possibility of a pipeline for the effective translation of basic science to new therapies, drug delivery, and personalized medicine.

Johnson and Picard. Advancing Neuroscience through Wearable Devices. Neuron (2020). Access the original scientific publication here.