Background Rosacea is a chronic inflammatory disease with variable clinical presentations, including transient flushing, fixed erythema, papules, pustules, and phymatous changes on the central face. Owing to the diversity in the clinical manifestations of rosacea, the lack of objective biochemical examinations, and nonspecificity in histopathological findings, accurate identification of rosacea is a big challenge. Artificial intelligence has emerged as a potential tool in the identification and evaluation of some skin diseases such as melanoma, basal cell carcinoma, and psoriasis. Objective The objective of our study was to utilize a convolutional neural network (CNN) to differentiate the clinical photos of patients with rosacea (taken from 3 different angles) from those of patients with other skin diseases such as acne, seborrheic dermatitis, and eczema that could be easily confused with rosacea. Methods In this study, 24,736 photos comprising of 18,647 photos of patients with rosacea and 6089 photos of patients with other skin diseases such as acne, facial seborrheic dermatitis, and eczema were included and analyzed by our CNN model based on ResNet-50. Results The CNN in our study achieved an overall accuracy and precision of 0.914 and 0.898, with an area under the receiver operating characteristic curve of 0.972 for the detection of rosacea. The accuracy of classifying 3 subtypes of rosacea, that is, erythematotelangiectatic rosacea, papulopustular rosacea, and phymatous rosacea was 83.9%, 74.3%, and 80.0%, respectively. Moreover, the accuracy and precision of our CNN to distinguish rosacea from acne reached 0.931 and 0.893, respectively. For the differentiation between rosacea, seborrheic dermatitis, and eczema, the overall accuracy of our CNN was 0.757 and the precision was 0.667. Finally, by comparing the CNN diagnosis with the diagnoses by dermatologists of different expertise levels, we found that our CNN system is capable of identifying rosacea with a performance superior to that of resident doctors or attending physicians and comparable to that of experienced dermatologists. Conclusions The findings of our study showed that by assessing clinical images, the CNN system in our study could identify rosacea with accuracy and precision comparable to that of an experienced dermatologist.

BACKGROUND Rosacea is a chronic inflammatory disease with variable clinical presentations including transient flushing, fixed erythema, papules, pustules and phymatous changes on the central face. Owing to the diversity of clinical manifestations, the lack of objective biochemical examinations and non-specificity of histopathology, accurate identification of rosacea is a big challenge. Artificial intelligence has emerged as a potential tool in the identification and evaluation of some skin diseases such as melanoma, basal cell carcinoma and psoriasis. OBJECTIVE In this work, we utilized convolution neural networks (CNN) to identify the clinical photos (from three different angles) of patients with rosacea and other diseases that would be easily confused with rosacea (such as acne, seborrheic dermatitis and eczema). METHODS In this work, we utilized convolution neural networks (CNN) to identify the clinical photos (from three different angles) of patients with rosacea and other diseases that would be easily confused with rosacea (such as acne, seborrheic dermatitis and eczema). RESULTS The CNN in our study achieved an overall accuracy and precision of 0.914 and 0.898, with an area under the receiver operating characteristic curve (AUROC) of 0.972 for the detection of rosacea. The accuracy of classifying the three subtypes of rosacea, ETR, PPR, PhR was 83.9%, 74.3% and 80.0%, respectively. Moreover, the accuracy and precision of our CNN to distinguish rosacea from acne reached 0.931 and 0.893, respectively. For the identificaiton between rosacea, seborrheic dermatitis and eczema, the overall accuracy was 0.757 and the precision was 0.667. Finally, by comparing the CNN with different levels of dermatologists, we showed that our CNN system is capable of identifying rosacea with a performance superior to resident doctors or attending physicians and comparable to experienced specialists. CONCLUSIONS In conclusion, by assessing clinical images, the CNN system in our study performed at dermatologist-level in the identification of rosacea. CLINICALTRIAL None

Objective To evaluate the epidemiology and disease burden of androgenetic alopecia (AGA) in college freshmen in China. Methods This population-based cross-sectional survey was carried out among 9227 freshmen of two comprehensive universities in two cities of China (Changsha and Xiamen) from September 2018 to October 2018. Questionnaires covering basic issues, surrounding demographic information, history of diseases, living habits, comorbidities, etc. were completed online in a self-reported manner Dermatological examination was performed by certified dermatologists. The disease burden of AGA, which includes health-related quality of life, symptoms of anxiety, symptoms of depression and quality of sleep, was measured by EQ-5D-3L, PHQ-2, GAD-2 and PSQI, respectively. Results The prevalence of AGA in college freshmen in China was 5.3/1000. Male was significantly associated with higher prevalence of AGA (7.9/1000, P<0.01) while female with lower risk of AGA (OR = 0.29, P = 0.002). There was no significant association between BMI and AGA, nor predilection of AGA in the Han nationality or the other ethnic minorities. Annual household income or parental highest educational level exerted no significant influence on the prevalence of AGA. Rosacea (OR = 3.22, P = 0.019) was significantly associated with higher prevalence of AGA while acne seemed not to be related to AGA. The scores of EQ-5D, GAD-2, PHQ-2 and PSQI were not significantly different between students with and without AGA. Conclusion The onset of AGA in Chinese college freshmen differ between genders and was significantly associated with rosacea.

BACKGROUND Androgenetic alopecia (AGA) is the most common type of hair loss and can onset at any age. It was reported to have a significant effect on health-related quality of life and psychological health. Reported prevalence rate is associated with multiple factors. However, these prevalence and disease burden in adolescents has not been explored. OBJECTIVE To evaluate the epidemiology and disease burden of AGA in Chinese adolescents. METHODS This population-based cross-sectional survey was carried out among 9227 freshmen of two comprehensive universities in two cities of China from September to October 2018. Dermatological examination was performed by certified dermatologists. Questionnaires covering basic issues, surrounding demographic information, history of diseases, etc. were completed online. Disease burden of AGA, including health-related quality of life, symptoms of anxiety and depression, and quality of sleep, was measured by EQ-5D-3L, PHQ-2, GAD-2 and PSQI. Meanwhile, 45 adolescents AGA patients visiting Xiangya hospital were enrolled as clinic AGA group for comparison of disease burden. RESULTS The prevalence of AGA in Chinese adolescents was 5.3/1000. Male was significantly associated with higher prevalence of AGA (7.9/1000, P<0.01). Female was associated with lower risk of AGA (OR=0.29, P=0.002). There was no significant association between BMI and AGA, nor predilection in different nationality. Annual household income or parental highest educational level exerted no influence on the prevalence of AGA. Rosacea (OR=3.22, P=0.019) was significantly associated with higher prevalence of AGA while acne seemed no association. The scores of EQ-5D, GAD-2, PHQ-2 and PSQI were not significantly different between students with and without AGA. However, the GAD-2, PHQ-2 and PSQI of clinic adolescent AGA patients were significantly higher than college patients. CONCLUSIONS The onset of AGA in Chinese college students differs between genders and was significantly associated with rosacea. Clinic AGA patients suffered from significant heavier disease burden compared to college AGA patients.