Related Experiment Video
Updated: Jun 19, 2026
Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
Published on: February 14, 2018
Optimization and Device-Level Validation of the i-Phagia Surface Electromyography Submental Sensor Patch for
Georgia A Malandraki1,2,3, Ching-Hsuan Peng1, Kevin Berry1
1Department of Speech, Language, and Hearing Sciences, Purdue University, West Lafayette, IN.
Purpose:
Surface electromyography (sEMG) is a promising biofeedback method for swallowing intervention. Current sensors are either costly rigid platforms or short-lived wearables. We developed a second-generation flexible wearable sEMG patch (i-Phagia) designed to record submental muscle activity. This study validated the signal quality and preclinical performance of i-Phagia against conventional sensors in two age groups.
Method:
A randomized crossover study involved 30 young (17 women, 13 men; aged 18-29 years) and 30 older healthy adults (17 women, 13 men; aged 50-82 years). Participants completed liquid and semisolid swallows under two conditions: with i-Phagia sensors and with conventional sensors, in a counterbalanced order. Comparisons focused on signal-based factors (signal-to-noise ratio [SNR], baseline amplitude, and normalized mean sEMG amplitude), as well as safety, satisfaction, and placement efficiency. SNR and baseline amplitude were examined with linear mixed models and equivalence tests; Pearson correlations were used to analyze normalized mean sEMG amplitude. Safety/adverse effects were assessed with descriptive statistics, while satisfaction and efficiency were analyzed with linear mixed models.
Results:
SNR values were equivalent between sensor types (effect size: -0.0178; confidence interval [CI] [-0.2226, 0.2004]); however, the i-Phagia patch showed significantly lower baseline amplitude than conventional sensors (effect size: 0.48; CI [0.3144, 0.6422]), indicating better signal quality. High correlations (r > .7) between normalized sEMG amplitudes across swallows suggest consistent muscle activity patterns across sensors. No pain was reported; skin redness was rare but twice as common after removing conventional sensors. Satisfaction was higher (p < .0001), and application was more efficient with i-Phagia (M = 1:28 vs. 2:19, p < .05).
Conclusions:
The optimized i-Phagia patch is safe, provides signal quality comparable to conventional sensors, and shows advantages in user satisfaction and placement efficiency. These findings support its feasibility for future clinical testing as a wearable sEMG option for swallowing rehabilitation.
Supplemental Material:
https://doi.org/10.23641/asha.32608575.
