Subscribe to eTOC

Electronic Sensors Break New Ground in Neurology Practice and Research



THE WRIST BAND device, being tested by Dr. Tobias Loddenkemper and colleagues to detect seizures, combines a three-dimensional accelerometer, which senses motion and velocity, and a skin conductivity sensor, which is sensitive to tiny amounts of sweat, indicating activation of the sympathetic nervous system.

Researchers are using wireless communication systems and electronic sensors in research aiming to improve detection and monitoring of seizures, movement disorders, and cognitive decline.

Imagine the possibilities: A real-time, wearable electronic sensor records a child's seizure, wirelessly sends an alert to the physician and caregiver, and delivers a dose of antiseizure medication. A motion sensor detects the walking cadence of a woman with advanced Parkinson's disease, adjusting the display of a moving grid pattern on the lenses of the Google glasses she is wearing, prompting her to step along the virtual tiles only she can see. And motion sensors strategically placed throughout the apartment of an elderly man, who had been diagnosed only four months earlier with depression and mild cognitive impairment at his twice yearly check-up, detect a rapid decline in his walking speed and his frequency of excursions outside the home, signaling his physician's office to schedule a visit three months early, where he is diagnosed and treated.

These applications are not that way off into the future. In ways, small and large, researchers are moving ahead to embrace wireless communication systems that connect more devices in the hands of neurology patients — and facilitate widespread integration of sensors, wearable computer computers, and other devices. These smaller, more portable systems portend future trends in the ways in which neurologic conditions are being assessed, monitored, and treated.

In interviews with Neurology Today, neurologists who are leading the charge on these innovations discussed the technological advances that they predict could transform important aspects of patient care.


Tobias Loddenkemper, MD, an associate professor of neurology at Harvard Medical School in Boston, is one of a growing number of epilepsy researchers developing wearable seizure-monitoring devices that can predict, record, and perhaps even treat seizures. “The problem we are trying to solve is twofold,” he said. “In the short term, we are hoping to detect acute seizure episodes. In the longer term, we want to better track seizure patterns, which can then ultimately be used to make better treatment decisions based on the data.”

“In the best-case scenario, the device detects and treats at the same time,” he said, with either an electrical stimulator or an implanted microcatheter delivery treatment. The technology is not there yet, but progress is being made.

In 2013, he noted, a research team from the University of Melbourne in Australia reported a “proof of concept” study in The Lancet Neurology that, in patients with frequent uncontrolled seizures, an implanted EEG monitor, wirelessly connected to a handheld computer, could learn the patient's seizure pattern and then alert the patient to an impending seizure.

The two-year study included 15 epilepsy patients between the ages of 20 and 62, who had experienced between two and 12 seizures a month and were refractory for treatment.

During the first month of the trial, the researchers used the system to record EEG data to enable them to develop algorithms to predict seizures for each patient. The system correctly predicted seizures 65 percent of the time, with a level better than 50 percent in 11 of the 15 patients. Of these 11, eight had their seizures accurately predicted between 56 and 100 percent of the time.

Dr. Loddenkemper is evaluating a wristband device to detect and record motor seizures. The device, which was developed in a lab at the Massachusetts Institute of Technology, combines a three-dimensional accelerometer, which senses motion and velocity — and a skin conductivity sensor, which is sensitive to tiny amounts of sweat, indicating activation of the sympathetic nervous system.


VIRTUAL REALITY goggles used in the study by Dr. Alberto Espay contain a built-in LCD screen, which projects floor tiles when subjects are moving, and earphones that sound step-matched cues as detected by the sensor strapped to their belts.

In 2012, he and his colleagues reported in Neurology that the device may be useful for predicting risk of sudden unexpected death in epilepsy (SUDEP). They used the wrist-worn sensors to continuously record the sympathetically mediated electrodermal activity (EDA) of patients with refractory epilepsy admitted to the long-term video-EEG monitoring unit. . They also measured from EKG recordings high-frequency (HF) heart rate variability, a measure of parasympathetic vagal modulation of the heart.

They analyzed 34 seizures comprising 22 complex partial and 12 tonic-clonic seizures from 11 patients, finding that the postictal period was characterized by a surge in EDA and heightened heart rate coinciding with persistent suppression of HF power.

The magnitude of both sympathetic activation and parasympathetic suppression increased with duration of EEG suppression after tonic-clonic seizures. Since decreased vagal activity is believed to be a risk factor for sudden cardiac death in other conditions, the results highlighted a critical window of postictal autonomic dysregulation that may be relevant in the pathogenesis of SUDEP, Dr. Loddenkemper and colleagues wrote.

In a 2012 study in Epilepsia, Dr. Loddenkemper used sensors to detect generalized tonic-clonic seizures, first training the device to recognize the combination of wrist motion and increased sweating that characterize such seizures, and then testing it in 80 patients over more than 4,000 hours of continuous activity. The device correctly detected 15 of 16 seizures, while registering 130 false alarms (throwing dice and playing Nintendo were the major culprits). Further training is likely to overcome this limitation, he said.


DR. TOBIAS LODDENKEMPER: “The problem we are trying to solve is twofold. In the short term, we are hoping to detect acute seizure episodes. In the longer term, we want to better track seizure patterns, which can then ultimately be used to make better treatment decisions based on the data.”

Major challenges remain before this or any other device becomes routine in clinical care of epilepsy, Dr. Loddenkemper cautioned. “We don't yet know what the right sensors are, or which is the best one for which type of patient.”

Regulatory hurdles may be significant, he noted, and none of the current generation of devices have been approved by the Food and Drug Administration. And despite the fact that real-time, home-based electronic monitoring is likely to be cost-effective compared with emergency room visits, there are currently no billing codes for home monitoring, making reimbursement problematic. Further trials, more development, and enhanced familiarity with wearable sensors are likely to increase the demand for such devices, and perhaps the willingness of insurance companies to pay for them, he said.

For now, while it is “still an unanswered question whether any of these devices improve care, in some ways, it is obvious: if you have more information, you may be able to treat better,” he said.


In other areas, movement disorders specialists are developing systems to prevent the gait freezing in advanced Parkinson's disease. There have been many attempts to create systems that provide patients with sensory stimuli that reset their stalled motor programs, including painting patterns on the floor of the home, and projecting laser light across the walking path from a specially designed cane.

Alberto Espay, MD, FAAN, an associate professor of neurology at the University of Cincinnati College of Medicine in Ohio, is experimenting with virtual reality goggles. The goggles are transparent, and have a small LCD screen embedded within them. An accelerometer, which the patient also wears, detects the patient's movements, triggering the display on the screen of floor tiles, moving toward the patient at a speed matching the patient's gait.

“The tiled floor acts as a moving visual display whose speed is generated in a natural feedback fashion by the patient's own motion, much like earth-stationary visual cues,” Dr. Espay explained. “Our experience has been that the patients benefit tremendously.”


DR. ALBERTO ESPAY is experimenting with virtual reality goggles to help ameliorate the incidence of gait freezing in advanced Parkinsons disease.

In a 2010 trial of the device, which also includes a set of earbuds that can provide auditory cues, which some patients prefer, Dr. Espay and colleagues reported that twice-daily use of the system for two weeks in 13 patients with off-state gait impairment increased walking speed from 62 centimeters per second to 73 cm/sec, and stride length from 74 cm to 84 cm. These effects were also seen just after removal of the device, suggesting some residual improvement. Freezing episodes showed a trend toward improvement, and United Parkinson's Disease rating scale motor scores also improved over the course of the training. The study was published in the Journal of Rehabilitation Research and Development.

Not every patient stands to benefit, however: those with freezing gait while in the “on” state seem resistant to this type of sensory cueing. “We think there is a different pathophysiology behind this phenomenon,” Dr. Espay said.

The device that Dr. Espay experimented with was a prototype, and it was not universally accepted by patients. Some found it bulky and socially awkward to use, he noted, adding:“Google Glass might be perfect for this application.”


Jeffrey Kaye, MD, FAAN, a professor of neurology and biomedical engineering at the Oregon Health and Science University in Portland, where he is also the director of the Layton Aging and Alzheimer's Disease Center, has developed an array of sensors that can be used throughout the home to monitor and report on physical, cognitive, and emotional health. The goal, he said, “is to provide timely and effective information that can prevent or delay the need for higher levels of care” — to allow seniors to age in their own communities.

Balance problems, cognitive impairment, and depression can all be detected with fairly simple sensors, when they are wired together and programmed appropriately. Passive, infrared motion detectors, typically one per room, plus a series in a hallway, can determine walking speed, movement between rooms, frequency of trips to the bathroom, and sleep patterns, as well as time spent out of the home, a decline in which is one sign of depression.


DR. JEFFREY KAYE: “We can, by using this technology, assess mobility, sleep, socialization, and cognitive function. When we put it all together, we can predict with a high level of sensitivity those people who are going to transition to needing more care within the next six months.”

“We can, by using this technology, assess mobility, sleep, socialization, and cognitive function,” Dr. Kaye said. “When we put it all together, we can predict with a high level of sensitivity those people who are going to transition to needing more care within the next six months.”

In 2012, Dr. Kaye and colleagues compared in-home walking speed between 31 patients with non-amnestic mild cognitive impairment (MCI) and 54 cognitively intact controls. They reported in Neurology that those with MCI not only walked more slowly, but the variability in their walking speed over time changed in important ways, and the combination of speed and variability allowed them to reliably distinguish those with mild cognitive impairment.

Specifically, while walking speed variability was moderate and stable in control subjects, it was either very high or very low in patients with MCI. The high variability, Dr. Kaye suggested, is likely reflecting early failure of gait regulation, while the low variability reflects late-stage failure, as overall functional ability declines.

The elderly people in his studies have been receptive to the data collection system, Dr. Kaye noted. “They perceive the benefit in something that might help them to stay in their home. They are much less resistant to this than to another clinic visit, in my experience,” he said.

Dr. Kaye foresees using such a system to provide more timely care, for instance prescribing physical therapy for balance problems or antidepressants for a mood change much earlier than these would occur from a reliance on clinical visits alone. He is currently involved in a three-year, randomized trial to determine whether in-home sensor systems can delay transition to nursing home placement or other skilled care.

The most immediate impact of the technology, he predicts, may be clinical trials, because of its ability to detect trends earlier. “You are gathering the information continuously, so you get a more accurate trajectory of change,” Dr. Kaye said. “There is a big opportunity to make a big impact with this type of approach.”

Several trends are driving the emergence of smart, networked sensors in medicine, including neurology, according to Wendy Nilsen, PhD, a Health Scientist Administrator at the NIH Office of Behavioral and Social Sciences Research.

First is the extraordinary computing power of mobile technology. “Your so-called ‘phone’ is really a massive computing device,” she pointed out, which, combined with extreme miniaturization and plummeting prices, allows such devices to be used in ways and in contexts never before possible. “We don't have to go to clinics for assessments in many cases,” she said. “So the question becomes, are there better ways to do these things?”


DR. WENDY NILSEN said that while the quality of current health management “apps” is highly variable, some are impressive, “and the quality is only going to go up. Well begin to see providers prescribing these.”

Second, she said, is the cost-driven need to do more with less, “and technology allows you to do that.” Third is the growing level of acceptance, from both patients and providers, of devolving more aspects of care to the patient or caregiver. Diabetes is the prime example at moment. “If you have diabetes, 99.9 percent of your management is in your hands,” she noted. While the quality of current health management “apps” is highly variable, some are impressive, “and the quality is only going to go up. We'll begin to see providers prescribing these.”

The new technology has raced ahead in part, she said, because it is drawing on a large set of experts, “including mobile network experts, gaming people, behavioral interaction specialists, and the traditional health community,” each contributing to the design of personalized devices to meet individual health needs. “The science of mobile health has brought together fields that traditionally didn't play together, all of them coming together in a way to change health.”



No caption available.

At press time, Apple had generated buzz around its announced launch of an open source software framework, called ResearchKit, which aims to help clinicians and scientists gather data from participants using iPhone apps. Scheduled for release in April, the ResearchKit will include activity modules that will provide information on memory or gait testing as well as other health measures. When granted permission by the user, the apps will be able to access data from the Health app on the iPhone on weight, blood pressure, glucose levels and asthma inhaler use.

The platform lets users choose the studies they want to join, and they can control the information they provide to specific apps and can see the data they are sharing. The user's personal data will be protected.

Currently, ResearchKit has apps in place for studies on asthma, breast cancer, cardiovascular disease, diabetes, and Parkinson's disease.

The app for Parkinson's disease, dubbed Parkinson mPower (Mobile Parkinson Observatory for Worldwide, Evidenced-based Research), was developed by Sage Bionetworks, a non-profit research organization based in Seattle, in partnership with the University of Rochester Medical Center neurologists Ray Dorsey, MD, MBA, a professor of neurology and co-director of the Center for Human Experimental Therapeutics, and Karl Kieburtz, MD, MPH, FAAN, director of the Center for Human Experimental Therapeutics and the Robert J. Joynt professor in neurology and professor of community and preventive Medicine and Environmental Medicine., and Max Little, PhD, a mathematician and lecturer at Aston University in the United Kingdom.

The mPower app, which is currently available for download through the Apple App store, will be used as a part of an observational study that collects data from Parkinson's patients on dexterity, balance and gait, voice and memory at multiple times each day.

For example, the app will measure dexterity by tracking how fast a person can tap the screen on their iPhone. Participants will be able to record their voice at different points during the day, enabling researchers to detect subtle changes, including tremor and reduced amplitude that will help assess any changes or severity of the symptoms. The device's GPS and accelerometer will measure mobility and balance. The Robert Wood Johnson Foundation is funding the study.

Dr. Dorsey and Sage's chief commons officer, John Wilbanks, discussed the app in an interview with Nature on March 10. Read the interview here at

Fay Ellis


•. Cook MJ, O'Brien TJ, Berkovic SF, et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study Lancet Neurol 2013; 12(6): 563–571.
    •. Poh MZ, Loddenkemper T, Reinsberger C, Swenson NC, Goyal S, Madsen JR, Picard RW. Autonomic changes with seizures correlate with postictal EEG suppression Neurology 2012;78(23):1868–1876.
      •. Poh MZ, Loddenkemper T, Reinsberger C, et al. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor Epilepsia 2012 May;53(5):e93–7
        •. Ramgopal S, Thome-Souza S, Jackson M, et al. T. Seizure detection, seizure prediction, and closed-loop warning systems in epilepsy Epilepsy Behav 2014;37:291–307
          •. Espay AJ, Baram Y, Dwivedi AK, et al. At-home training with closed-loop augmented-reality cueing device for improving gait in patients with Parkinson disease J Rehabil Res Dev 2010;47(6):573–581.
            •. Kaye J, Mattek N, Dodge HH, et al. Unobtrusive measurement of daily computer use to detect mild cognitive impairment Alzheimers Dement 2014; 10(1)
              •. Dodge HH, Mattek NC, Austin D, et al. In-home walking speeds and variability trajectories associated with mild cognitive impairment Neurology 2012;78(24):1946–1952