My data is currently in the form of multiple image files (png), all within a single folder. There are a range of associated metadata contained in the file name. For example, the format of each file name is something like 'patientId_dateOfBirth_sex_modality_date_time.png'. I would like to analyze this data by importing it into a tibble in R and then manipulating it to make it tidy.

I have come across a few suggestions which use the Command Prompt but I was hoping for a solution using R script to make it reproducible. I think that if I can export the file names into a tibble then I should be able to figure out how to make it tidy using stringr, but I'm not sure what sort of import options are available.

Thank you!

CodePudding user response：

Some actual example filenames would have been helpful to understand the problem better.

However, you may use list.files to list all the files in a particular directory and since all the values are separated by _ you may use separate to split the values in different columns.

library(dplyr)
library(tidyr)

filenames <- list.files('/path/to/png/files')

#Used these two filenames as example
#filenames <- c("A123_19910622_M_2_20230114_042312.png", 
#               "A128_19910828_F_4_20221214_142110.png")

tibble(filenames) %>%
  separate(filenames, c('patientId', 'dateOfBirth', 'sex', 'modality', 
                        'date', 'time'), extra = "drop")

# A tibble: 2 × 6
#  patientId dateOfBirth sex   modality date     time  
#  <chr>     <chr>       <chr> <chr>    <chr>    <chr> 
#1 A123      19910622    M     2        20230114 042312
#2 A128      19910828    F     4        20221214 142110