ABSTRACT
In today抯 scenarios many healthcare decisions are being taken by predictive modeling and machine learning techniques. With this review, we focused on logistic regression model, a kind of predictive modeling used in machine learning, and how healthcare researchers take decisions by the help of predictive modeling. For a better data analysis in healthcare, we need to understand the concept of logistic regression as well as others terms, which are linked with it. so that we can clearly understand the concept behind it and implement in medical research. In this review we worked on an example and illustrated how to perform logistic regression using R programming language. The aim of this paper is to understand logistic regression in healthcare and implement it for decision making.
ABSTRACT
【Objective】 To compare the performance of five commonly used variable selection methods in high-dimensional biomedical data variable screening so as to explore the effects of sample size and association among candidate variables on screening results and provide evidence for the development of variable selection strategy in high-dimensional biomedical data analysis. 【Methods】 Variable selection algorithms were implemented based on R-programming language. Monte Carlo method was used to simulate high-dimensional biomedical data under different conditions to evaluate and compare the performance of different variable selection methods. Variable selection performance was evaluated based on the true positive rate and true negative rate in screening. 【Results】 For specified high-dimensional data, the variable selection performance was improved for all the methods when sample size was increased, and the association between candidate variables did affect variable screening results. Simulation results indicated that the elastic network algorithm yielded the best screening performance, LASSO algorithm took the second place, and ridge algorithm did not work at all. 【Conclusion】 Elastic network algorithm is an ideal variable screening method for high-dimensional data variable screening.
ABSTRACT
Objective To explore the effect of age on the fracture healing through bioinformatical analysis of gene ex?pression data in GEO, and to screen critical molecular targets and pathways involved in this process. Methods Through R programming language, we identified different expressed genes between 26/52 week old rats and 6 week old rats in every time points of the experiment (No fracture;3 days, 1 week, 2 weeks, 4 weeks and 6 weeks after fracture). By comparison of these different expressed genes, those genes that may contribute to fracture healing were identified. Function annotation was conducted based on DAVID database and PPI network that was constructed via STRING database. Results Compared with 6 week old rat, 52 week old rat show more different genes at 2, 4 and 6 weeks after fracture as well as more than intact rats. At the time point of 6 weeks after fracture, 26 week old rat present 4 different genes while 52 week old rat present 99 differ?ent genes compared with 6 week old rat. We totally found 99 genes that might play important roles in the process of fracture healing. These genes involved in biological process related to bone healing, immune, inflammatory and etc. Also, two screened gene enriched KEGG pathways were identified: ECM-receptor interaction and Arachidonic acid metabolism. Through the analysis of PPI network, Pcna, Fn1, Casp3 and etc, who present high density connectivity in PPI network, were screened out. Conclusion Pcna, Casp3 and Fn1 and etc might play important roles in fracture healing through affecting ECM-receptor interaction and Arachidonic acid metabolism.