Sharp Brains: Brain Fitness and Cognitive Health News

Neuroplasticity, Brain Fitness and Cognitive Health News


Next in clinical practice: Automated real-time detection of seizures via wearable EMG devices

(A) The wear­able device placed on the brachial biceps mus­cles. (B, C) The wear­able device, which is con­nect­ed to the self-adhe­sive patch, con­tain­ing the record­ing elec­trodes and the ground elec­trode. (D) Remote con­trol of the device. (Epilep­tic seizure Detec­tor Devel­oped by Ictal­Care). Cred­it: Neu­rol­o­gy.


Wear­able EMG Found to Detect Seizures (Neu­rol­o­gy Today):

A new study demon­strates the fea­si­bil­i­ty of using a wear­able elec­tromyo­g­ra­phy device to detect ton­ic-clonic seizures…The Neu­rol­o­gy paper was among the first to demon­strate its results prospec­tive­ly, using a pre-spec­i­fied cut-off for deter­min­ing that a GTCS is occur­ring. And at nine sec­onds, its laten­cy in doing so (from the time of onset as mea­sured by an inde­pen­dent observ­er) is also among the fastest described so far, the study authors and inde­pen­dent experts not­ed…

The new paper comes on the heels of two oth­er stud­ies on wear­able EMG for seizure detection…Altogether, they demon­strate the grow­ing fea­si­bil­i­ty of incor­po­rat­ing wear­able seizure-detec­tion devices into clin­i­cal prac­tice — both to quan­ti­fy for physi­cians how well patients are respond­ing to med­ica­tion, and to help care­givers respond quick­ly to ongo­ing seizures.

With a sen­si­tiv­i­ty of 93.8 per­cent, the device detect­ed 30 of 32 GTCS over a total record­ing time of 3,735.5 hours. The rate of false alarms was 0.67 per day, but two-thirds of the patients had no false alarms. For 24 of 71 patients (34 per­cent) who did expe­ri­ence a false alarm, the most com­mon rea­son was due to phys­i­cal exer­cise, which account­ed for 68 per­cent of all false alarms. Only two false alarms occurred dur­ing sleep.

Dr. Beniczky said the over­all rate of 0.67 false alarms per day “still needs to be improved. It’s not such a prob­lem for patients, because it hap­pens almost entire­ly dur­ing the day, so they can stop the alarm when it hap­pens. But it is not good enough if we want a tru­ly auto­mat­ic device.”

The Study

Auto­mat­ed real-time detec­tion of ton­ic-clonic seizures using a wear­able EMG device (Neu­rol­o­gy). From the abstract:

  • OBJECTIVE: To deter­mine the accu­ra­cy of auto­mat­ed detec­tion of gen­er­al­ized ton­ic-clonic seizures (GTCS) using a wear­able sur­face EMG device.
  • METHODS: 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.
  • RESULTS: 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.
  • CONCLUSIONS: 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.
  • CLASSIFICATION OF EVIDENCE: 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).

The Study in Context

Leave a Reply...

Loading Facebook Comments ...

Leave a Reply

Categories: Cognitive Neuroscience, Technology

Tags: , , , , , , , , , , , ,