Automatic Monitoring of High-Frequency Autoimmune Disorder Related Seizures with Wearable Devices

Area of Science:

• Neurology
• Biomedical Engineering
• Wearable Technology

Background:

• Faciobrachial Dystonic Seizures (FBDS) in LGI1-IgG autoimmune encephalitis are difficult to detect via EEG.
• High seizure frequency and lack of clear EEG signals pose diagnostic challenges.

Purpose of the Study:

• To evaluate the feasibility of using wrist-worn wearable devices for noninvasive FBDS detection.
• To develop an objective seizure diary using wearable technology.
• To differentiate FBDS from normal nocturnal arousals.

Main Methods:

• Seven LGI1-IgG patients and four controls wore Empatica wristbands in hospital and ambulatory settings.
• A two-stage semi-supervised learning approach with a Support Vector Machine (SVM) classifier was employed.
• Wearable signal characteristics were analyzed for differences before and after immunotherapy.

Main Results:

• Wearable devices detected significant differences in signal characteristics for FBDS compared to controls.
• Pre-treatment FBDS events were more frequent, longer, and associated with increased Electrodermal Activity (EDA).
• Significant differences in wearable signal characteristics were observed before and after immunotherapy.

Conclusions:

• Wearable technology shows potential for automatic and objective FBDS measurement, particularly during sleep.
• This approach minimizes patient burden compared to traditional seizure diaries.
• Wearable devices can aid in monitoring treatment efficacy for LGI1-IgG autoimmune encephalitis.