Related Experiment Video
Updated: May 24, 2026

Dynamic Digital Biomarkers of Motor and Cognitive Function in Parkinson's Disease
Published on: July 24, 2019
Recognition of Microexpressions Using Nonlinearly Transformed Zygomaticus Muscle Activation
Sarva Kiruthika Aruldass1, Karthick P A2, Ramakrishnan Swaminathan1
1Non-Invasive Imaging and Diagnostics Lab, Department of Applied Mechanics and Biomedical Engineering, IIT Madras, Chennai, India.
Abstract:
This study classifies happy and sad Micro-Expressions (MEs) based on facial Electromyography (fEMG) signals recorded from the zygomaticus region. A muscle activation technique, which is typically used to model the nonlinear EMG-force relationship, is utilized to examine ME signals across varying nonlinear shape factor A values. Time-Domain (TD) features extracted from muscle activation signals were statistically analyzed, and training a Random Forest (RF) classifier achieved the highest accuracy of 91.6%. In contrast, accuracy decreases when features are derived from the normalized signal. These results demonstrate that muscle activation enhances emotion classification accuracy.
More Related Videos
Related Concept Videos
Muscles for Facial Expressions
Facial Feedback Hypothesis
Nonconscious Mimicry
Assessment of Airway, Skin Color, and Use of Accessory Muscles
Introduction
The initial evaluation of a patient's respiratory system...
Non-Verbal Cues

