Data Mining and Meta-Analysis of Psoriasis Based on Association Rules

Area of Science:

• Dermatology
• Pharmacology
• Data Mining

Background:

• Psoriasis vulgaris is a chronic dermatological condition significantly impacting patient well-being.
• Evaluating treatment efficacy and safety is crucial for managing psoriasis vulgaris.
• Existing data mining algorithms like Apriori have limitations in efficiency and data handling.

Purpose of the Study:

• To evaluate the effectiveness and safety of topical defubao for psoriasis vulgaris through meta-analysis.
• To introduce a novel matrix association rule method based on random forest weighting for improved data analysis.
• To compare defubao with calcipotriol and betamethasone in treating psoriasis vulgaris.

Main Methods:

• Included 11 studies evaluating topical defubao for psoriasis vulgaris.
• Utilized efficacy indicators: Investigator's Global Assessment (IGA), Patient's Global Assessment (PGA), and Psoriasis Area and Severity Index (PASI).
• Employed a matrix association rule method with random forest weighting for data analysis and efficiency.

Main Results:

• Defubao demonstrated superior treatment effects compared to calcipotriol and betamethasone, with statistically significant differences (IGA, PGA).
• An 8-week defubao treatment duration yielded significantly better outcomes than 4-week or 2-week durations.
• Defubao treatment resulted in a higher average PASI reduction rate after 4 weeks and a significantly lower incidence of adverse events after 8 weeks compared to calcipotriol.

Conclusions:

• Topical defubao is an effective and safe treatment for psoriasis vulgaris.
• Defubao shows significant advantages over calcipotriol and betamethasone in efficacy and safety.
• Longer treatment durations (8 weeks) enhance defubao's therapeutic benefits and safety profile.