With Artificial Intelligence (AI) revolutionizing the field of aesthetic medicine, we are beginning to see innovative approaches to patient care and treatment. With continued advancements, AI technology offers the potential to deliver more accurate and objective assessments in aesthetic medicine, paving the way for a new standard in clinical evaluations.
Recognizing this transformative potential of AI, a group of experts from the aesthetic and dermatology communities recently came together to establish a set of guidelines that would harness the power of AI in assessing baseline appearance and treatment outcomes. While traditional methods utilizing validated scales have been the norm, we are now entering an era where AI can help overcome the limitations of human bias and variability often associated with these methods. This post will discuss their findings, exploring how AI technology can help shape the future of aesthetic medicine.
Consensus Aims and Methodology
The consensus group, composed of experts from multiple specialties such as plastic surgery and dermatology, convened to establish guidelines and recommendations for the development, implementation, and application of AI in aesthetic medicine.
To set a foundation for these discussions, a pre-survey questionnaire was distributed to participants, covering the following topics:
- Facial examination strategies
- AI assistance in aesthetic medicine
- Objectifying facial examination
- Clinical steps improvement
- Skin quality assessment
- Holistic facial indices
- Root-specific assessment and treatment
- Root-guided AI algorithm refinement
- Research and future directions
- Ethical considerations
- AI application usability
- Expanding scope
The results of this pre-consensus survey revealed that for most practitioners, the use of AI in patient analysis was limited, with more focus being put on aligning treatment plans with patient preferences. Despite this, they recognized AI’s potential for improving treatment monitoring, identifying options, and boosting efficiency. Objectification tools in facial exams, particularly photographic documentation, were revealed as the most highly valued for enhancing patient care, as was provider experience. The survey also demonstrated that participants see AI as playing a key role in consultations and follow-ups, especially when it comes to gender-specific insights and the development of new facial indices. Furthermore, there is a growing interest in AI that can adapt to a variety of patient needs and it was recognized that more research is needed to optimize its application in patient assessments and personalized treatment plans.
Consensus Meeting Results
The below, taken directly from the consensus meeting publication, lists the statements that received a 100% agreement of the participants involved.
- AI implemented in aesthetic medicine can help to standardize and improve patient assessment and patient consultation.
- AI implemented in aesthetic medicine can help to prevent overcorrection.
- There is a need for validated objective facial assessments in aesthetic medicine.
- Facial aesthetic assessment should be comprehensive and be based on objective indices (as FAI, FYI or SQI).
- The skin quality assessment should differentiate between female and male skin.
- Patients whose skin is covered with make-up must be excluded from AI examination at baseline.
- Patients’ age and gender should be included in the AI assessment.
- Patients’ ancestral roots should be included in the AI system.
Analysis of Ancestral Roots
Beyond the core objectives, the team of experts also discussed the importance of incorporating ancestral roots into AI assessments, ensuring that key characteristics of various ethnic groups are recognized and considered.
They conducted a comprehensive analysis of facial features for various ethnic groups, including East Asian, Indian, European, Latin and African. As a result, they set out a detailed analysis of each of these groups, which is represented below:
East Asian – monolid eyes, square lower face shapes, broader midfaces, a characteristic flat and short nose, and a typically retruded chin.
Indian – prominent forehead, brown to black hair, and eye color, as well as significantly larger eyes, well-defined tear troughs and a retruded chin.
European – gender-differentiated eyebrow shapes, with women displaying a peaked brow and men a more horizontal brow line, convex forehead, high zygomatic arches, thinner lips, and an inclination towards static wrinkles as a sign of aging.
Latin – skin tone spectrum from Fitzpatrick Skin Type II–V and varied eye coloration, ranging from light to very dark brown. Smaller foreheads and a bizygomatic distance equal to or smaller than the bigonial distance were typical, width of the lips was noted to be more than 20% of the total lower facial width.
African – darker skin tones with a Fitzpatrick Skin Type >III, may have noses shorter than one third of the face length, but broader than 1/5 of the facial width, lips broader than one third of the lower facial third, as well as a forehead area broader than one third of the facial length, oftentimes shows a neutral or positive canthal tilt, even in elderly patients.
Consensus Outcomes and Discussion
The consensus revealed that the integration of AI into aesthetic medicine is viewed as pivotal for standardizing patient assessments and consultations and improving consistency and objectivity across the field. Where traditionally, evaluations have been subjective, relying on expert judgements, AI can help to optimize these evaluations while also mitigating variability and reducing the risk of overcorrection which is seen as a common issue in aesthetic procedures.
They further highlight the importance of adopting validated facial assessment indices such as the Facial Aesthetic Index (FAI), Facial Youth Index (FYI), and Skin Quality Index (SQI). It is believed these can help quantify youthful and healthy skin attributes, balanced facial features, and overall aesthetic appeal.
The consensus also emphasized the importance of incorporating gender-specific, age-based, and culturally sensitive AI assessments in aesthetic medicine. It is recognized that biological differences between male and female skin, such as thickness and aging patterns, necessitate tailored treatment plans. AI systems with gender-specific and age-specific algorithms can offer more precise recommendations for each group.
Providing personalized care and treatment also necessitates the inclusion of ancestral roots to account for the diverse populations. “An AI system that incorporates ancestral data can better understand and predict individual responses to treatment, align with patients’ aesthetic ideals within their roots, and maintain the authenticity of ethnically specific features.”
However, to address any complexities associated with the absence of a definitive cutoff for distinguishing between the ancestries, the consensus group has made the following recommendations. AI models should be trained in ethical and cultural sensitivity to reflect the global population, accounting for mixed ancestries. They should also allow for a case-by-case approach to categorization while acknowledging globalization’s influence on beauty standards and preserving ethnic and cultural nuances. The guidelines also emphasize the need for ethical considerations to be incorporated into AI models to ensure they are free from bias and do not perpetuate a narrow definition of beauty. Lastly, it is recommended that an interdisciplinary collaboration be taken across fields such as cosmetic surgery, dermatology, and anthropology to ensure a holistic and ethically sound approach.
Concluding Thoughts
This consensus on AI in aesthetic medicine has underscored its transformative potential in standardizing assessments and enhancing treatment efficacy to elevate patient care. As the field continues to evolve, these guidelines represent a meaningful step forward, paving the way for more precise, personalized, and ethically sound approaches in aesthetic medicine.
Sources:
- Frank, K., Day, D., Few, J., Chiranjiv, C., Gold, M., Sattler, S., Kerscher, M., Knoedler, L., Filippo, A., Rzany, B., Cotofana, S., Fabi, S., Fritz, K., Peng, P., Wanitphakdeedecha, R., Pooth, R., & Huang, P. (2024). AI assistance in aesthetic medicine–A consensus on objective medical standards. Journal of Cosmetic Dermatology. https://doi.org/10.1111/jocd.16481
