setwd(\'D:\\SCIwork\\F4\\clinical\')

require(table1)

rm(list=ls()) 

library(dplyr)

library(tidyr)

data <- read.csv(\'survival.csv\', header = T, row.names = 1) 

library(stringr)

colnames(data)

data <- data[-(1:6), ]

data$T  <-   str_extract(data$TNM, "T\\d")


data$Te1  <-   str_extract(data$TNM, "T\\d[a-z]+")
data$Te1 <- ifelse(is.na(data$Te1),data$Te2,data$Te1)


data$N  <-   str_extract(data$TNM, "N\\d")
data$N  <-   str_extract(data$N, "\\d")

data$M <-   str_extract(data$TNM, "M\\d")
data$M <-   str_extract(data$M, "\\d")



data$stage1  <- str_trim(str_extract(data$Stage, "\\s[H-Z]+"),
                         side = c("both", "left", "right"))

data$stage <- ifelse(data$stage1 == "X" ,\'\', data$stage1)


data$stage 

colnames(data)

data$Age <- ifelse(data$Age > 60, \'> 60\',\'<= 60\')

table(data$Age)

table(data$OS)



## 对我们感兴趣的变量因子化
data$OS <- 
  factor(data$OS, 
         levels=c(0,1),
         labels=c("Alive", # 第一个作为参考组
                  "Death"))
table <- table1(~ factor(Gender) + factor(Age) + factor(stage) + 
         factor(T) +  factor(N) +  factor(M) | OS, data=data)

table