Innovating Efficacy Testing for Anti-aging Cosmetic Products by Wrinkle Grading with AI-Powered SCAWA Scales
In recent years, advances in artificial intelligence (AI) have permeated various industries, offering innovative solutions to age-old challenges. The field of cosmetology – and wrinkle grading in particular – is no exception. Enter the Standardized ColorFace® AI-based Wrinkle Assessment (SCAWA) Scale – a new digital grading tool, recently published in the International Journal of Cosmetic Science (Rengot et al, 2024), designed to simplify and enhance the clinical grading of wrinkles.
This digital solution addresses industry challenges related to clinical grading by generating standardized and realistic wrinkle representations through machine learning technology. Using generative adversarial networks (GANs), SCAWA creates artificial yet accurate images to help dermatologists, cosmetic researchers and manufacturers evaluate the efficacy of anti-aging products with highest accuracy, consistency and convenience.
Share This Research Highlight
Using Wrinkle Grading in Evaluating Anti-Aging Product Efficacy
The Need for Accurate Wrinkle Grading
Dermo-cosmetic companies aim to reduce wrinkles, but proving the effectiveness of their products is challenging. Solutions that promise to reduce or delay the appearance of wrinkles must be backed by scientific data.
But achieving objective, repeatable results in clinical evaluations is no small feat. Traditional wrinkle evaluation methods rely on visual assessment by trained experts using grading tools based on real photographs. Standardized photographic scales representative of different wrinkle depths are used to assign reliable grades to subjects. These real-world image-based tools typically exist as hard copies (printed books or sheets) for in vivo grading.
With the rise of digital technology, there’s been a push toward more standardized, digital, and data-driven approaches.
A New Online Grading Tool to Help Graders Better Assess Wrinkle Severity
The evaluation of anti-aging treatments is often based on visual wrinkle grading, in which clinical experts refer to photographic scales based on real photographs. Rengot et al. used an entirely new approach based on artificial intelligence (machine learning) to complement assessments made using photographic atlases and address some of the challenges often associated with their use.
Using computer-generated images rather than real photographs, the authors wanted the new scale to:
- Be mono-parametric (morphologically stable)
- Capture the full range of wrinkle depths
- Show well-detailed wrinkle evolution (mathematically linear scale)
- Be easy to use
The SCAWA Scale is intended to be an innovative digital alternative that allows both photographic and real-world grading to evaluate and substantiate product efficacy claims.
Standardizing Wrinkle Grading with the SCAWA Scales
Wrinkle Grading enhanced with AI and machine learning models
AI and the revolutionary Generative Adversarial Networks (GANs) – these machine learning models consist of two neural networks in a zero-sum game, creating synthetic yet realistic images. While traditionally used to generate full-face images, GANs are now being used to represent specific skin features.
However, generating high-resolution images remains a challenge. While technology has advanced to produce images up to 1024×1024 pixels, generating realistic images focused on human skin remains complex. This underscores the innovation that drives projects like SCAWA.
The creation of the SCAWA scales required a robust data set, beginning with the acquisition of high-resolution photographs. Digital portraits of 587 Caucasian women between the ages of 18 and 70 were taken in three European cities. The diverse wrinkle appearance of the participants ensured a representative dataset. Data augmentation techniques were used to increase the robustness of the dataset. By applying random deformations, the researchers artificially expanded their dataset and optimized model performance. Pre-processing steps, including color image registration and cropping, further refined the dataset.
Using GANs, SCAWA creates high-resolution synthetic images that accurately represent varying wrinkle depths in a linear and standardized manner.
Standardizing Wrinkle Assessment – How the SCAWA Scales Work
SCAWA scales relies on standardization of wrinkle grading, here’s how the technology works:
- Digital Wrinkle Simulation: GANs generate artificial, photorealistic wrinkle images by training on existing data and refining the output using a base image.
- Morphology control: Generated images are fine-tuned to represent specific wrinkle characteristics without introducing extraneous variables such as skin tone or facial features.
- Linearity and precision: Grade increments are optimized to ensure small, mathematically consistent differences between successive wrinkle representations.
The result? A 12-point crow’s-feet wrinkle scale that is realistic, reliable and easy to use.
Figure 1: SCAWA scales for crow’s feet evaluation
The entire scale can be adjusted to highlight a single feature, such as wrinkles. In addition, artificial images allow fine-tuning of all desired aspects of the scale using precise morphological criteria. Integration of the SCAWA scales with Photoscale®, a complete online image evaluation tool, allows skin to be conveniently evaluated on photographs.
How SCAWA Scales Can Improve Cosmetic Product Testing
The authors challenged the SCAWA scales in a study, and the key findings and results show:
- Realism and robustness: SCAWA scales produce lifelike wrinkle images that have high coherence and correlate strongly (R=0.94) with the well-known reference Skin Aging Atlas (Bazin and Doublet, 2007) in grading assessments.
- Improved Consistency: AI-generated scales reduce inter-rater variability and provide greater homogeneity between scores by minimizing the impact of irrelevant details (e.g., different morphologies).
- Improved accessibility: Unlike photographic atlases, which are limited by privacy regulations (e.g., GDPR), AI-generated images bypass legal restrictions, making them more practical for global use.
- Cost and efficiency gains: No need to recruit subjects or conduct new photo shoots. Once trained, GANs can produce virtually unlimited scales tailored to specific needs.
These results validate the scale’s reliability in standard use cases and highlight its potential for widespread adoption in cosmetic product development.
Benefits for the Cosmetic Industry
SCAWA scales offer several important benefits to researchers, dermatologists and cosmetic companies:
- Standardization: Consistent wrinkle grading eliminates bias and ensures that assessments are accurate, repeatable and objective.
- Customization: Scales can be tailored to different demographics, wrinkle types, and product evaluation goals.
- Efficient product testing: Fast and inexpensive image generation enables large-scale comparative testing across geographies.
- Global usability: Digital formats facilitate remote grading, allowing researchers worldwide to evaluate standardized images.
These advances are particularly relevant to the evolving landscape of cosmetic efficacy testing, where transparency, traceability and rapid innovation are critical.
Practical Implications for Cosmetologists
SCAWA scales are an invaluable tool for cosmetologists. By providing a consistent, objective method of wrinkle grading, these scales enable professionals to more accurately assess product efficacy.
This, in turn, increases client confidence and satisfaction, resulting in positive outcomes for both clients and practitioners.
Using the SCAWA Scales for Wrinkle Grading in Clinical Evaluations
The SCAWA scales also hold great promise for clinical evaluations. By providing a standardized, AI-driven method of wrinkle assessment, these scales ensure consistent results across different populations and geographies. This harmonization is critical for multi-center studies where variability can skew results.
For dermatologists and researchers, this tool offers a new level of precision and accessibility, paving the way for more comprehensive studies and better patient outcomes.
Leveraging SCAWA for the development of anti-aging product
The SCAWA scale represents a paradigm shift for the dermo-cosmetics industry, reinforcing its commitment to science-based, consumer-centric solutions. By leveraging AI to optimize wrinkle grading, brands can
- Back up anti-aging claims with rigor and transparency.
- Improve product development cycles by accurately measuring treatment efficacy.
- Increase consumer confidence with robust, data-driven assessments.
By providing a reliable way to assess product efficacy, these scales enable companies to refine their formulations and optimize their offerings. This, in turn, increases market competitiveness and builds consumer confidence.
Future Development – Expanding the Application of SCAWA Scales to Support Anti-Aging Claims
While the initial SCAWA initiative focused on crow’s feet wrinkles in Caucasian subjects, plans are already underway to expand the application:
- Diverse demographics: Create scales representing different ethnicities, genders, and age groups to ensure inclusivity in assessments.
- Additional signs of aging: Developing scales for other signs of aging, such as nasolabial folds and glabellar lines.
- Optimized systems: Integrate enhanced features such as improved brightness controls and intermediate grading options for more nuanced analysis.
- Expanded Device Compatibility: Adapt SCAWA scales for use with multiple imaging devices to replicate diverse real-world conditions.
These efforts are aimed at refining the utility of SCAWA tools to ensure they remain at the forefront of innovation in wrinkle assessment and claim substantiation.
Interested in the SCAWA scales technology?
Contact our experts to discuss the capabilities and applications for your project.
Discover our articles
Learn more about our
Imaging Systems
The ColorFace®
The DynaCam®
Watch Our Expert Present
The SkinCam Pro®
Download our brochure on
Skin Longevity & Aging
Watch our Webinars
References
Bazin R, Doublet E. Skin aging atlas. Paris: Med’com 2007.
