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

Jiarui Ou,Jianglin Zhang,Enas Abdulhay

doi:10.1155/2022/9188553

Abstract

Psoriasis is a common chronic and recurrent disease in dermatology, which has a great impact on the physical and mental health of patients. Meta-analysis can evaluate the effectiveness and safety of defubao in the treatment of psoriasis vulgaris. This article observes psoriasis skin lesions treated with topical defubao and the changes in blood vessels under dermoscopy. Considering that the Apriori algorithm and the existing improved algorithm have the problems of ignoring the weight and repeatedly scanning the database, this paper proposes a matrix association rule method based on random forest weighting. This method uses the random forest algorithm to assign weights to each item in the data set, and introduces matrix theory to convert the transaction data set into a matrix form and store it, thereby improving operating efficiency. This article included 11 studies, of which 7 studies used the indicator “Researcher's Overall Assessment” (IGA) to evaluate the efficacy, 5 studies used the “Patient Overall Assessment” (PGA) as the efficacy evaluation index, and Loss Area and Severity Index (PASI) was used as an observation index to evaluate the efficacy. Seven studies conducted safety comparisons. In this paper, IGA and PGA were used as evaluation indicators. The treatment effect of the defubao group was better than the calcipotriol group and the betamethasone group. The differences were statistically significant. The effect of the Fubao treatment for 8 weeks is significantly better than that of 4 weeks and 2 weeks, and the differences are statistically different. Using PASI as the evaluation index, a descriptive study was carried out, and it was found that after 4 weeks of treatment for psoriasis vulgaris, the average PASI reduction rate of patients was higher than that of the calcipotriol group and the betamethasone group. The safety evaluation found that after 8 weeks of treatment, the incidence of adverse events in the defubao group was significantly lower than that in the calcipotriol group.

Highlights

How to discover these hidden and valuable relationships has always been a research hotspot in the field of data mining [1, 2]
Data mining technology refers to a large number of data sets, through association rule mining, classification, aggregation and outlier detection and other methods, to find the internal relationship between the data in the data set without obvious characteristics, so as to guide people in subsequent production activities
In the subgroup comparing the efficacy of 4 weeks, the Cochrane Q analysis suggests that the combined data of the three groups are significantly heterogeneous, so the 4-week period is valid. e conclusions reached suggest that after 4 weeks of treatment, the treatment effect of the defubao group is better than that of the betamethasone group

Summary

Introduction

How to discover these hidden and valuable relationships has always been a research hotspot in the field of data mining [1, 2]. Data mining technology refers to a large number of data sets, through association rule mining, classification, aggregation and outlier detection and other methods, to find the internal relationship between the data in the data set without obvious characteristics, so as to guide people in subsequent production activities. Is mining technology has been widely used in many fields such as store layout, consumer behavior, accident data analysis, and fraud detection [6, 7]. Known as “white boil” in traditional Chinese medicine, is a common chronic, inflammatory, and recurrent skin disease [8]. It is characterized by red papules or plaques with multiple layers of silvery white scales covering the limbs. Our country is in a period of social transformation and rapid economic growth, and the subsequent air pollution, water pollution and increasing living pressure have laid hidden dangers for the occurrence of this disease [13]

Methods

Results

Conclusion