diff --git a/readme.md b/readme.md
index 8db41d0..41c9fc5 100644
--- a/readme.md
+++ b/readme.md
@@ -88,7 +88,10 @@ A typical data directory might look like:
     |           |-- GT                                      <-- The corresponding ground truth folder for the .nii file  
                     | -- name.nii or name.nii.gz            <-- The ground truth (GT) image in nifti format (Name can be anything)
 
-### ⚙️Installation
+
+ **Note:** the folder name for PET images should be `PET` and for the ground truth `GT`. All other folder and subfolder names could be anything. 
+ 
+ ### ⚙️Installation
 
 Please read the documentation before opening an issue!
 
@@ -192,11 +195,11 @@ model is designed in easy configurable mode.
   Kindly check if these generated files are in order.
   
   
-- Predicted results, including sTMTV and sDmax, will be saved into the folder `output/*.csv`, 
-where the CSV file's name is automatically generated with the time of prediction in the month, year, and time.
+- Predicted results including predicted segmentation masks and calculated surrogate biomarekrs (sTMTV and sDmax) will be saved into the folder `output/.*`. 
 
-
-- Surrogate biomarkers (sTMTV and sDmax) will be automatically calculated and saved as an EXCEL file inside the folder `output/*.csv`
+ - Predicted masks are saved under the folder name `output/predicted/.*`. The predicted masks are indicated with the keyword `predicted` in the file name, and the input PET images are indicated with the keyword `pet`. For example, `patient_id_predicted.nii` for predicted mask of `patient_id` and `patient_id_pet.nii` for the PET images. If the ground truths are given, they are saved with the name `patient_id_ground_truth.nii`. Each .nii file has both sagittal and coronal views Each .nii file has both sagittal and coronal views concatenated.  
+ 
+- Surrogate biomarkers (sTMTV and sDmax) will be automatically calculated and saved as an EXCEL file inside the folder output/*.csv. Two EXCEL files will be saved. The first one constitutes computed surrogate biomarkers calculated from the segmentation masks predicted from AI with an indicator `predicted` in the file name. The second EXCEL file would constitute the surrogate biomarkers computed from the reference segmentation masks (i.e., ground truth) from the expert (if available) with an indicator `ground_truth` in the file name. In addition to the `predicted` and `ground truth` indicator names, the CSV file's name also constitutes an automatically generated month, year, and the processing time.
 
 
 ### 🙋 FAQ