Accuracy and clinical relevance of an automated, algorithm‐based analysis of facial signs from selfie images of women in the United States of various ages, ancestries and phototypes: A cross‐sectional observational study

Abstract

Background: Real-life validation is necessary to ensure our artificial intelligence (AI) skin diagnostic tool is inclusive across a diverse and representative US population of various ages, ancestries and skin phototypes.Objectives: To explore the relevance and accuracy of an automated, algorithm-based analysis of facial signs in representative women of different ancestries, ages and phototypes, living in the same country. Methods:In a cross-sectional study of selfie images of 1041 US women, algorithmbased analyses of seven facial signs were automatically graded by an AI-based algorithm and by 50 US dermatologists of various profiles (age, gender, ancestry, geographical location). For automated analysis and dermatologist assessment, the same referential skin atlas was used to standardize the grading scales. The average values and their variability were compared with respect to age, ancestry and phototype.Results: For five signs, the grading obtained by the automated system were strongly correlated with dermatologists' assessments (r ≥ 0.75); cheek skin pores were moderately correlated (r = 0.63) and pigmentation signs, especially for the darkest skin tones, were weakly correlated (r = 0.40) to the dermatologist assessments. Age and ancestry had no effect on the correlations. In many cases, the automated system performed better than the dermatologist-assessed clinical grading due to 0.3-0.5 grading unit differences among the dermatologist panel that were not related to any individual characteristic (e.g. gender, age, ancestry, location). The use of phototypes, as discontinuous categorical variables, is likely a limiting factor in the assessments of grading, whether obtained by automated analysis or clinical assessment of the images. Conclusions:The AI-based automatic procedure is accurate and clinically relevant for analysing facial signs in a diverse and inclusive population of US women, as confirmed by a diverse panel of dermatologists, although skin tone requires further improvement.

Objective and automatic grading system of facial signs from smartphones’ pictures in South African men: Validation versus dermatologists and characterization of changes with age

“…However, some coefficients are of lower values in some facial signs (cheek skin pores, marionette lines), possibly due to impact of vellus hairs. Same holds true for some age‐classes where coefficients appear lower, suggesting necessary improvements of the system, as previously discussed 18 . Creation of an accurate, inclusive, and effective AI‐based automatic grading system in men facial skin, 8 and especially for dark phototypes groups, is instrumental to fill knowledge need gaps as in literature these panels are underrepresented.…”

“…Same holds true for some age-classes where coefficients appear lower, suggesting necessary improvements of the system, as previously discussed. 18 Creation of an accurate, inclusive, and effective AI-based automatic grading system in men facial skin, 8 and especially for dark phototypes groups, is instrumental to fill knowledge need gaps as in literature these panels are underrepresented. Impacts of solar radiations, chronic urban pollution exposures, or stress factors/lifestyles could be studied in fast and remote ways leveraging owners' smartphones and associated AI-based algorithms, as previously done in women.…”

“…8 This highlights the importance of using well-distributed datasets across all demographics from the algorithmic evaluation. Clearly, the different training datasets used in the development on new AI-based automated grading system for men and women [7][8][9]18 fulfill these requirements by covering different geographical locations (USA, Brazil, India, China, Korea, Japan, South Africa, France) with different ages, phototypes, gender, and ancestries. To continue to enrich the automatic grading When comparing data obtained with AI-based automated grading system in Chinese men 8 till 50−59 years, wrinkles/texture, pigmentation, and ptosis/sagging scores appear lower, whereas cheek pores seem close.…”

“…An AI‐based automated grading system was developed and validated to assess facial signs, 7–10 with an accuracy based upon dermatologists’ gradings using Skin Aging Atlas photographic standardized scales as references 11–17 . The system has been developed using selfie images collected in different countries to ensure accuracy for different genders, skin tones, and ancestries 7–9,18 …”

“…[11][12][13][14][15][16][17] The system has been developed using selfie images collected in different countries to ensure accuracy for different genders, skin tones, and ancestries. [7][8][9]18 Such technology enables scientists to conduct, in a fast and agile way, studies that cannot be imagined before, 19 and fill the knowledge gaps in different populations like in emerging markets or in men populations. 8,20 Indeed, numerous studies and knowledge already exists in North hemisphere in phototypes I-IV in female panels, but embracing diversity with new accurate data in different gender and in darkest phototypes are mandatory.…”