Facial acne intelligent grading method and system based on image analysis
Image analysis technology enables refined identification and quantitative assessment of facial acne, solving the problems of imprecise lesion identification, inconsistent assessment, and lack of quantitative efficacy in existing technologies, and providing accurate assessment of acne severity and monitoring of treatment effects.
Patent Information
- Authority / Receiving Office
- CN Β· China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE FIRST AFFILIATED HOSPITAL OF WENZHOU MEDICAL UNIV
- Filing Date
- 2026-05-06
- Publication Date
- 2026-07-24
AI Technical Summary
Existing methods for classifying facial acne suffer from insufficient granularity in lesion identification, poor consistency in assessments among different physicians, and a lack of objective quantitative indicators for treatment efficacy, resulting in imprecise, subjective, and unquantifiable assessments.
An intelligent grading method for facial acne based on image analysis is adopted, including standardized acquisition of facial images and skin color normalization preprocessing, localization of facial key points and resolution of regions of interest, segmentation and automatic counting of multiple types of acne lesions, region-weighted multi-feature fusion scoring, and image registration and efficacy quantitative evaluation before and after treatment, to construct a complete closed-loop solution.
It enables precise identification and automatic counting of skin lesion types such as acne, papules, pustules, nodules, and cysts, improving the clinical interpretability and consistency of the assessment results. It also provides an objective efficacy assessment by calculating the lesion clearance rate and the number of new lesions through image registration before and after treatment.
Smart Images

Figure CN122453783A_ABST
Abstract
Citation Information
Patent Citations
Facial acne grading method based on collaborative mapping multi-scale self-distillation
CN119599992A