SCAWA scales: A new digital tool for wrinkles clinical grading based on AI

Area of Science:

• Dermatology and cosmetic science
• Artificial intelligence in medical imaging
• Computer vision for aesthetic assessment

Background:

• Clinical wrinkle depth assessment is crucial for anti-aging product efficacy.
• Current photographic scales are often hard-copy based, limiting accessibility.
• Digital tools are needed for improved grading comfort and flexibility.

Purpose of the Study:

• To develop a digital, AI-based standardized scale for wrinkle assessment.
• To create computer-generated wrinkle images for grading scales.
• To enhance the accuracy, linearity, and accessibility of wrinkle evaluation tools.

Main Methods:

• Utilized a generative adversarial network (GAN) to create realistic, controllable wrinkle images.
• Trained the GAN to generate a 12-point scale for crow's feet wrinkles.
• Selected artificial images for morphological stability and mathematical linearity.

Main Results:

• The AI-based scale (SCAWA) is realistic, linear, and accurate for photographic assessments.
• Validated scale coherence, reliability, and acceptability in real-world use.
• Achieved high correlation (R=0.94) with existing assessment methods, improving expert harmonization.

Conclusions:

• The SCAWA scale leverages machine learning for an innovative digital wrinkle assessment tool.
• Optimized linearity, homogeneity, and accuracy in visual grading.
• Positive expert feedback suggests potential for broader application in cosmetic efficacy testing.