Sunscreen UV Protection Prediction Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting UV protection performance of sunscreen products are limited by the accuracy of UV combination-based models, which fail to account for formulation ingredients like emollients and texture types, leading to discrepancies between in vivo and in vitro test results.
Innovation Solution
A method and system using machine learning techniques to predict UV protection by selecting features such as viscosity, emollient polarity, UV filter types, and their ratios, and inputting these into a predictive model for calculating UV protection values, optionally employing dimensionality reduction techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If UV combination based models are used to predict UV protection performance, then the prediction process is simplified, but the prediction accuracy is limited because the models fail to account for formulation ingredients such as emollients, emulsifiers, and polymers
Solution Approach 1:
The patent transforms the prediction approach by changing the parameters fed into the machine learning model. Instead of only UV filter combinations, the model now receives multiple formulation parameters including emollient types (high, medium, low polarity), emulsifiers, polymers, and texture types. This parameter expansion enables the model to capture the complex interactions between ingredients while maintaining a unified prediction framework.
Solution Approach 2:
The patent treats the sunscreen formulation as a composite system where multiple ingredients (UV filters, emollients, emulsifiers, polymers) interact together. The machine learning model is trained to recognize these composite interactions, particularly how emollient polarity and texture types modify the UV protection performance of filter combinations, thereby improving prediction accuracy through composite ingredient analysis.
2Productivity
If in vitro tests are used to evaluate UV protection performance, then the testing process is faster and less costly, but there are still gaps between in vitro and in vivo values for both SPF and UVA-PF
Solution Approach 1:
The patent introduces a machine learning prediction model as an intermediary between in vitro measurements and in vivo performance. The model is trained on datasets containing both in vitro test results and corresponding in vivo outcomes, learning the complex relationships between formulation composition, in vitro performance, and actual in vivo protection. This intermediary model can then predict in vivo-like performance from in vitro data, bridging the accuracy gap while maintaining testing efficiency.
3Measurement precision
If multiple formulation parameters including emollients, emulsifiers, and polymers are included in the predictive model, then the prediction accuracy improves, but the model complexity and data processing requirements increase
Solution Approach 1:
The patent segments the formulation parameters into distinct functional categories: UV filters (with sub-categories of absorbing and scattering types), emollients (segmented by polarity: high, medium, low), emulsifiers, and polymers. Each category is processed and evaluated separately in the machine learning model, allowing for systematic analysis of ingredient contributions while managing model complexity through structured data organization.
Data Source
AI summary
The present invention relates to a method and a system for UV protection prediction of a sunscreen product, comprising a) selecting the features of the sunscreen product, wherein the features include a viscosity, a high polarity emollient, a medium polarity emollient, a low polarity emollient, a UVA filter, a UVB filter, a ratio of UVB filter vs UVA filter, a ratio of UV filter in oil phase vs UV filter in water phase, and a ratio of absorbing type UV filter vs scattering/reflecting type UV filter; c) inputting the features into a predictive model, which is built and fitted by using one or more machine learning techniques; and d) calculating the UV protection prediction value of the sunscreen product by the predictive model of step c).