Abstract:The study assessed the correspondence in live conditions between clinical gradings by dermatologists and an AI‐based algorithm analysing selfie images for facial aging signs in 125 Chinese subjects. Results showed high correlation with clinical assessments, identifying areas for improvement in the AI system. Objective The objective of this study is to assess the correspondence, in live conditions, between clinical gradings of facial aging signs by three dermatologists and those afforded by an automatic AI‐based algorithm that analyses smartphones' selfie images of Chinese subjects. Methods In total, 125 Chinese subjects of both genders, aged 18–62y, took a selfie using their own smartphones and were immediately viewed by three dermatologists. The latter graded the severity of 15 facial signs in women and 9 in men, using the standardized values afforded by a Skin Aging Atlas referential dedicated to Asian skin. The data issued by both methodologies were then statistically compared. Results The absolute gradings of the automatic system were found highly correlated with clinical assessments, with lower values in most cases. In women, large differences in absolute values were found on the gradings for size of isolated spot, cheek fold, spread macules, and texture of mouth contour women. Analysis of the Mean Absolute Errors (M.A.E) revealed that these rarely exceed 0.6 grading units in women and to a lesser extent in men. Conclusion The present study confirmed the value of the automatic system towards an extended use towards large human cohorts as a surrogate of clinical evaluations and allowed to detect the points where improvements must be brought to the system. Résumé Objectif évaluer la correspondance, en conditions réelles, entre les classements cliniques des signes de vieillissement du visage par trois dermatologues et ceux établis par un algorithme automatique d'IA qui analyse les images de selfie sur smartphone de sujets chinois. Méthodes 125 sujets chinois des deux sexes, âgés de 18 à 62 ans, ont pris un selfie à l'aide de leurs smartphones et ont été immédiatement examinés par trois dermatologues. Ces derniers ont évalué la sévérité de 15 signes faciaux chez les femmes et 9 chez les hommes, en utilisant les valeurs standardisées établies par un référentiel de l'Atlas du vieillissement cutané dédié à la peau asiatique. Les données obtenues par les deux méthodologies ont ensuite été comparées statistiquement. Résultats les classements absolus du système automatique étaient fortement corrélés aux évaluations cliniques, avec des valeurs plus faibles dans la plupart des cas. Chez les femmes, des différences importantes en termes de valeurs absolues ont été observées concernant les classements de la taille de la tache isolée, du pli de la joue, des macules diffuses et de la texture du contour de la bouche chez les femmes. L'analyse des erreurs absolues moyennes (EAM) variables a révélé que celles‐ci dépassent rarement 0,6 unité de classement chez les femmes et, dans une moindre mesure, chez les hommes. Conclusion la présente étude a confirmé la valeur du système automatique en vue d'une utilisation étendue à de grandes cohortes humaines en remplacement des évaluations cliniques et a permis de détecter les points où des améliorations doivent être apportées au système.

Comparisons between Caucasian‐validated photo‐numeric scales and Korean‐validated photo‐numeric scales for photo‐ageing. Insights from the Singapore/Malaysia cross‐sectional genetics epidemiology study (SMCGES) cohort

Comparisons between Caucasian-validated and Chinese-validated photo-numeric scales for assessing facial wrinkles

Comparisons between wrinkles and photo-ageing detected and self-reported by the participant or identified by trained assessors reveal insights from Chinese individuals in the Singapore/Malaysia Cross-sectional Genetics Epidemiology Study (SMCGES) cohort

Comparisons between eyebags, droopy eyelids, and eyebrow positioning identified by photo‐numeric scales or identified by written descriptive scales: Insights from the Singapore/Malaysia cross‐sectional genetics epidemiology study (SMCGES) cohort

Classification of facial wrinkles among Chinese women

Survey on Skin Aging Status and Related Influential Factors in Southeast China

A Correlation of the Glogau Scale With VISIA-CR Complexion Analysis Measurements in Assessing Facial Photoaging for Clinical Research.

Reference Values for the Chinese Population of Skin Autofluorescence As a Marker of Advanced Glycation End Products Accumulated in Tissue

Comparison of facial skin ageing in healthy Asian and Caucasian females quantified by in vivo line-field confocal optical coherence tomography 3D imaging

Development and validation of a global photonumeric scale for evaluating skin quality of aged female facial skin

Changes in facial signs due to age and their respective weights on the perception of age, on a tired-look or a healthy glow among differently aged Chinese men

Modeling of global skin aging indices among Caucasian and Asian women

Genome-wide Scan Identified Genetic Variants Associated with Skin Aging in a Chinese Female Population

Knowledge, attitude, and practice toward photoaging in the Chinese population: a cross-sectional study

A new Chinese crow's feet grading scale based on the DermaTOP system

Extrinsic Skin Ageing in German, Chinese and Japanese Women Manifests Differently in All Three Groups Depending on Ethnic Background, Age and Anatomical Site

Assessing the reliability of four severity scales depicting skin ageing features

Grading facial aging: Comparing the clinical assessments made by three dermatologists with those obtained by an AI‐based scoring system that analyses selfie pictures. A focus on Chinese subjects of both genders

Skin Color Independent Assessment of Aging Using Skin Autofluorescence.

Structural and functional age-related changes in some facial signs of Chinese men: A pilot study

The continuous development of a complete and objective automatic grading system of facial signs from selfie pictures: Asian validation study and application to women of three ethnic origins, differently aged