Related Experiment Video
Updated: Mar 31, 2026
05:04
Facial Nerve Surgery in the Rat Model to Study Axonal Inhibition and Regeneration
Published on: May 5, 2020
8.2K
A smartphone-based automatic diagnosis system for facial nerve palsy
Hyun Seok Kim1, So Young Kim2, Young Ho Kim3
1Interdisciplinary Program of Bioengineering, Seoul National University, Seoul 03080, Korea. khs0330kr@bmsil.snu.ac.kr.
Sensors (Basel, Switzerland)
|October 28, 2015
Summary
This study introduces a smartphone system for diagnosing facial nerve palsy. The portable tool achieved 88.9% accuracy in distinguishing patients from healthy individuals using facial movement analysis.
Area of Science:
- Biomedical Engineering
- Medical Diagnostics
- Computer Vision
Background:
- Facial nerve palsy (FNP) results in facial muscle weakness due to facial nerve damage.
- Objective and reliable assessment tools are crucial for FNP diagnosis and management.
- Current assessment methods can be subjective and lack portability.
Purpose of the Study:
- To develop and validate a rapid, portable, smartphone-based system for automatic FNP diagnosis.
- To differentiate individuals with FNP from healthy controls using quantitative facial analysis.
Main Methods:
- Facial landmarks were tracked using an incremental parallel cascade of linear regression.
- An asymmetry index was calculated based on facial region displacement ratios during specific movements (resting, eyebrow raising, smiling).
- Classification was performed using Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) with Leave-one-out Cross-Validation (LOOCV) on 36 subjects.
Main Results:
- The developed system successfully discriminated between subjects with and without FNP.
- An overall classification accuracy rate of 88.9% was achieved.
- The system demonstrated the potential for quantitative and objective FNP assessment.
Conclusions:
- A smartphone-based system offers a feasible and accurate approach for FNP diagnosis.
- This technology can provide a portable and accessible tool for clinicians and patients.
- Further research can explore larger cohorts and diverse FNP etiologies.
