Calculated a pooled AUC, bootstrapped 95% CI and then thresholding the curve following MICE imputation

Question

I have been doing some fairly simple ROC curve analysis, involving creating some ROC curves, calculating AUC and 95% CI (2000 bootstrapped replicates) and then thresholding the curve to give a 95% sensitivity, with the threshold, TN, FN, FP, TP, Sens and Spec returned. This is simple when using a single dataset as below:

BIOMroc1 = pROC::roc(response = DF$CT, predictor = DF$Biom.1, ci=TRUE)
pROC::auc(BIOMroc1 ) 
pROC::ci.auc(BIOMroc1 , conf.level=0.95, method=c("bootstrap"), boot.n = 2000) 

However I am now wanting to use a estimated variable as the predictor that has required imputation to form, meaning there are 10 imputed datasets. I wish to calculate the AUC and 95% CI as similar to above and then combine them. Following this I'd like to somehow create a ROC curve and threshold at 95% sensitivity. I am abit lost as to where to start! Code for a replica dataset is below.

set.seed(123)

num_imputations <- 5

num_rows <- 20

generate_data <- function(num_rows, mean_range, sd_range) {
  data <- data.frame(
    CT = sample(0:1, num_rows, replace = TRUE),
    matrix(runif(num_rows * 6, min = mean_range[1], max = mean_range[2]), ncol = 6)
  )
  colnames(data)[-1] <- paste0("Biomarkers_", 1:6)
  return(data)
}

imputed_datasets <- lapply(1:num_imputations, function(i) {
  generate_data(num_rows, c(5, 10), c(2, 4))
})

imputed_datasets <- lapply(1:num_imputations, function(i) {
  data <- imputed_datasets[[i]]
  data$Imputation <- i
  return(data)
})

combined_data <- do.call(rbind, imputed_datasets)