Project-MONAI#413 add a function to calculate task params for DynUNet pipeline (Project-MONAI#414)

* add a function to calculate task params

Signed-off-by: Yiheng Wang <[email protected]>

* update readme

Signed-off-by: Yiheng Wang <[email protected]>

* update readme

Signed-off-by: Yiheng Wang <[email protected]>

Author: yiheng-wang-nv

modules/dynunet_pipeline/README.md

@@ -14,6 +14,14 @@ My running environment:
 
 To prevent the inconsistency, all json files are included in `config/` already.
 
+## Task parameters
+All task specific parameters refer from `nnUNet` pipeline (as shown in `task_params.py`).
+During these parameters, `patch_size`, `batch_size` and `deep_supr_num` are achieved from some heuristic rules, and this part is not covered in this pipeline. `spacing`, `clip_values` and `normalize_values` are based on the statistics of the dataset (only on training set). Here we provide `calculate_task_params.py` and you can try to calculate them. The command is like:
+
+```
+python calculate_task_params.py -task_id 09
+```
+
 ## Training
 Please run `train.py` for training. Please modify the command line arguments according
 to the actual situation, such as `determinism_flag` for deterministic training, `amp` for automatic mixed precision.

modules/dynunet_pipeline/calculate_task_params.py

@@ -0,0 +1,84 @@
+import os
+from argparse import ArgumentDefaultsHelpFormatter, ArgumentParser
+
+from monai.data import (
+    Dataset,
+    DatasetSummary,
+    load_decathlon_datalist,
+    load_decathlon_properties,
+)
+from monai.transforms import LoadImaged
+
+from task_params import task_name
+
+
+def get_task_params(args):
+    """
+    This function is used to achieve the spacings of decathlon dataset.
+    In addition, for CT images (task 03, 06, 07, 08, 09 and 10), this function
+    also prints the mean and std values (used for normalization), and the min (0.5 percentile)
+    and max(99.5 percentile) values (used for clip).
+
+    """
+    task_id = args.task_id
+    root_dir = args.root_dir
+    datalist_path = args.datalist_path
+    dataset_path = os.path.join(root_dir, task_name[task_id])
+    datalist_name = "dataset_task{}.json".format(task_id)
+
+    # get all training data
+    datalist = load_decathlon_datalist(
+        os.path.join(datalist_path, datalist_name), True, "training", dataset_path
+    )
+
+    # get modality info.
+    properties = load_decathlon_properties(
+        os.path.join(datalist_path, datalist_name), "modality"
+    )
+
+    dataset = Dataset(
+        data=datalist,
+        transform=LoadImaged(keys=["image", "label"]),
+    )
+
+    calculator = DatasetSummary(dataset, num_workers=4)
+    target_spacing = calculator.get_target_spacing()
+    print("spacing: ", target_spacing)
+    if properties["modality"]["0"] == "CT":
+        print("CT input, calculate statistics:")
+        calculator.calculate_statistics()
+        print("mean: ", calculator.data_mean, " std: ", calculator.data_std)
+        calculator.calculate_percentiles(
+            sampling_flag=True, interval=10, min_percentile=0.5, max_percentile=99.5
+        )
+        print(
+            "min: ",
+            calculator.data_min_percentile,
+            " max: ",
+            calculator.data_max_percentile,
+        )
+    else:
+        print("non CT input, skip calculating.")
+
+
+if __name__ == "__main__":
+    parser = ArgumentParser(formatter_class=ArgumentDefaultsHelpFormatter)
+    parser.add_argument(
+        "-task_id", "--task_id", type=str, default="04", help="task 01 to 10"
+    )
+    parser.add_argument(
+        "-root_dir",
+        "--root_dir",
+        type=str,
+        default="/workspace/data/medical/",
+        help="dataset path",
+    )
+    parser.add_argument(
+        "-datalist_path",
+        "--datalist_path",
+        type=str,
+        default="config/",
+    )
+
+    args = parser.parse_args()
+    get_task_params(args)