Related Experiment Video
Updated: Jun 11, 2025
10:28
Dynamic Digital Biomarkers of Motor and Cognitive Function in Parkinson's Disease
Published on: July 24, 2019
15.1K
Evaluating the Impact of BoNT-A Injections on Facial Expressions: A Deep Learning Analysis
Aesthetic Surgery Journal
|October 4, 2024
Summary
Botulinum toxin type A (BoNT-A) injections reduce the recognition of angry and surprised facial expressions. Deep learning objectively measured these changes, showing potential for a more positive appearance.
Area of Science:
- Aesthetics and Facial Rejuvenation
- Artificial Intelligence in Medicine
- Psychology and Emotion Recognition
Background:
- Botulinum toxin type A (BoNT-A) injections are common for facial rejuvenation.
- The precise effects of BoNT-A on facial expressions are not well understood.
Purpose of the Study:
- To objectively quantify the impact of BoNT-A injections on facial expressions.
- Utilize deep learning techniques for precise measurement.
Main Methods:
- 180 patients (25-60 years) received upper face BoNT-A injections.
- Photographs of neutral, happy, surprised, and angry expressions were taken pre- and post-injection.
- A convolutional neural network (CNN)-based facial emotion recognition (FER) system analyzed 1440 images.
Main Results:
- The CNN model achieved 90.15% accuracy in image prediction.
- Post-injection, recognition of angry and surprised expressions significantly decreased (P < .05).
- Angry expressions were misclassified as neutral/happy, and surprised as neutral (P < .05).
Conclusions:
- Deep learning offers a standardized method to assess BoNT-A's effect on facial expressions.
- BoNT-A may diminish anger and surprise expressions, potentially enhancing facial appearance.
- Further research is required to explore the broader implications of these findings.
