# Copyright (c) MONAI Consortium
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import numpy as np
import torch
from monai.transforms import LoadImage
from monailabel.interfaces.datastore import Datastore, DefaultLabelTag
from monailabel.interfaces.tasks.scoring import ScoringMethod
logger = logging.getLogger(__name__)
[docs]class Dice(ScoringMethod):
"""
Compute dice between final vs original tags
"""
def __init__(self):
super().__init__("Compute Dice for predicated label vs submitted")
def __call__(self, request, datastore: Datastore):
loader = LoadImage(image_only=True)
tag_y = request.get("y", DefaultLabelTag.FINAL)
tag_y_pred = request.get("y_pred", DefaultLabelTag.ORIGINAL)
result = {}
for image_id in datastore.list_images():
y_i = datastore.get_label_by_image_id(image_id, tag_y) if tag_y else None
y_pred_i = datastore.get_label_by_image_id(image_id, tag_y_pred) if tag_y_pred else None
if y_i and y_pred_i:
y = loader(datastore.get_label_uri(y_i, tag_y))
y_pred = loader(datastore.get_label_uri(y_pred_i, tag_y_pred))
y = y.flatten()
if isinstance(y, torch.Tensor):
y = y.numpy()
y_pred = y_pred.flatten()
if isinstance(y_pred, torch.Tensor):
y_pred = y_pred.numpy()
union = np.sum(y) + np.sum(y_pred)
dice = 2.0 * np.sum(y * y_pred) / union if union != 0 else 1
logger.info(f"Dice Score for {image_id} is {dice}")
datastore.update_image_info(image_id, {"dice": dice})
result[image_id] = dice
return result