This function takes a probability map for a single subject and predicts the subject-specific
threshold to apply based on the TAPAS model generated from tapas_train().
The function will return a list of objects including the TAPAS predicted subject-specific threshold,
the lesion mask produced from applying this threshold, and the lesion mask
produced from using the group threshold.
tapas_predict(pmap, model, clamp = TRUE, k = 0, verbose = TRUE)
| pmap | A |
|---|---|
| model | The TAPAS model fit from |
| clamp | A |
| k | The minimum number of voxels for a cluster/component. Segmentation clusters of size less than k are removed from the mask, volume estimation, and the Sørensen's–Dice coefficient (DSC) calculation. |
| verbose | A |
A list containing the TAPAS predicted subject-specific threshold (subject_threshold), the lesion
segmentation mask obtained using the TAPAS predicted subject-specific threshold (tapas_binary_mask), and the
lesion segmentation mask obtained using the group threshold (group_binary_mask).
if (FALSE) { # Data is provided in the rtapas package as arrays. Below we will convert them to nifti objects. # Before we can implement the train_tapas function we have to generate the training data library(oro.nifti) # Create a list of gold standard manual segmentation train_gold_standard_masks = list(gs1 = gs1, gs2 = gs2, gs3 = gs3, gs4 = gs4, gs5 = gs5, gs6 = gs6, gs7 = gs7, gs8 = gs8, gs9 = gs9, gs10 = gs10) # Convert the gold standard masks to nifti objects train_gold_standard_masks = lapply(train_gold_standard_masks, oro.nifti::nifti) # Make a list of the training probability maps train_probability_maps = list(pmap1 = pmap1, pmap2 = pmap2, pmap3 = pmap3, pmap4 = pmap4, pmap5 = pmap5, pmap6 = pmap6, pmap7 = pmap7, pmap8 = pmap8, pmap9 = pmap9, pmap10 = pmap10) # Convert the probability maps to nifti objects train_probability_maps = lapply(train_probability_maps, oro.nifti::nifti) # Make a list of the brain masks train_brain_masks = list(brain_mask1 = brain_mask, brain_mask2 = brain_mask, brain_mask3 = brain_mask, brain_mask4 = brain_mask, brain_mask5 = brain_mask, brain_mask6 = brain_mask, brain_mask7 = brain_mask, brain_mask8 = brain_mask, brain_mask9 = brain_mask, brain_mask10 = brain_mask) # Convert the brain masks to nifti objects train_brain_masks = lapply(train_brain_masks, oro.nifti::nifti) # Specify training IDs train_ids = paste0('subject_', 1:length(train_gold_standard_masks)) # The function below runs on 2 cores. Be sure your machine has 2 cores available or switch to 1. # Run tapas_data_par function # You can also use the tapas_data function and generate each subjects data data = tapas_data_par(cores = 2, thresholds = seq(from = 0, to = 1, by = 0.01), pmap = train_probability_maps, gold_standard = train_gold_standard_masks, mask = train_brain_masks, k = 0, subject_id = train_ids, ret = TRUE, outfile = NULL, verbose = TRUE) # We can now implement the train_tapas function using the data from tapas_data_par tapas_model = tapas_train(data = train_data1, dsc_cutoff = 0.03, verbose = TRUE) # Load a subject reserved for testing and convert to a nifti # Probability map pmap11 = oro.nifti::nifti(pmap11) # Brain mask brain_mask = oro.nifti::nifti(brain_mask) # Use TAPAS to predict on a new subject test_subject_prediction = tapas_predict(pmap = pmap11, model = tapas_model, clamp = TRUE, k = 0, verbose = TRUE) # Show subject-specific TAPAS threshold test_subject_prediction$subject_threshold # Look at TAPAS binary segmentation from applying the TAPAS threshold oro.nifti::image(test_subject_prediction$tapas_binary_mask) # Look at group threshold binary segmentation from applying the group threshold oro.nifti::image(test_subject_prediction$group_binary_threshold) }