Updated to version 2.1.0

README.md

@@ -72,6 +72,10 @@ This python version is differ than the originally published MATLAB version [http
 - In detection post-processing, a morphological erosion step follows the opening has been added with kernel radius of 5 pixels.
 - The 95th and 5th percentile of the fat and muscle intensities, respectively, were used to normalize the image.
 
+**`VERSION 2.1.0 and above`**
+- In case of non-detected fat or muscle objects, the 3 middel slices will be selected, and the normalization will be done using the intensity be calculated using the 95th and 5th percentile of the entire pre-prccessed 3 slices.
+
+
 # How to cite AutoRef/pyAutoRef
 In case of using or refering to AutoRef/pyAutoRef, please cite it as:
 ```

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = pyAutoRef
-version = 2.0.2
+version = 2.1.0
 author = Mohammed R. S. Sunoqrot
 author_email = [email protected]
 description = Python pakage to perfom AutoRef (prostate T2w MRI dual reference tissue [fat and muscle] normalization).

src/pyAutoRef/autoref.py

@@ -7,7 +7,7 @@
 from pyAutoRef.object_detection import object_detection
 from pyAutoRef.post_processing import post_process_predictions
 from pyAutoRef.normalization import normalize_image
-from pyAutoRef.utils import save_image, check_predictions, check_input_image
+from pyAutoRef.utils import save_image, check_predictions, check_input_image, get_intensities_without_detection
 
 """
 This is the python version of the:
@@ -83,12 +83,13 @@ def autoref(input_image, output_image_path=None):
         # Check again after recalculating
         class_with_zero_predictions = check_predictions(top_predictions)
 
-    # If still any class has zero predictions, raise an error
+    # If still any class has zero predictions, normalize without detection
     if class_with_zero_predictions:
-        raise ValueError(f"No detected objects for {class_with_zero_predictions}.")  
-
-    # Perform post-processing to the detected objects
-    processed_images_intensities = post_process_predictions(
+        print(f"Objects still not detected. Recalculating intensities without detection...")
+        processed_images_intensities = get_intensities_without_detection(resized_corrected_image)
+    else:
+        # Perform post-processing to the detected objects
+        processed_images_intensities = post_process_predictions(
         resized_corrected_image, top_predictions)
 
     # Perform normalization

src/pyAutoRef/utils.py

@@ -942,3 +942,49 @@ def check_input_image(input_image):
         return 'Path'
     else:
         raise ValueError("The entered value is not a SimpleITK.Image or a valid path.")
+
+def get_intensities_without_detection(image_3D):
+    """
+    Process the 3D image to get the intensities within the selected 3 middle slices.
+
+    Parameters:
+        image_3D (SimpleITK.Image): The 3D image containing the slices.
+
+    Returns:
+        processed_images_intensities (numpy.ndarray): An array contains all the intensites
+          within the selected 3 middle slices.
+    """
+    # Detect and remove outliers in the 3D image
+    image_3D_wo_outliers = detect_remove_outliers(image_3D)
+
+    # Get the size of the image
+    size = image_3D_wo_outliers.GetSize()
+
+    # Determine the central slices
+    central_slice_indices = [size[2] // 2 - 1, size[2] // 2, size[2] // 2 + 1]
+
+    # Initialize an empty list to store non-zero intensities
+    non_zero_intensities = []
+
+    # Loop through the central slices and extract non-zero intensities
+    for index in central_slice_indices:
+        # Extract the 2D slice
+        slice_image = image_3D_wo_outliers[:, :, index]
+        # Convert to numpy array
+        slice_array = sitk.GetArrayFromImage(slice_image)
+        # Get non-zero intensities
+        non_zero_values = slice_array[slice_array > 0]
+        # Append the non-zero values to the list
+        non_zero_intensities.extend(non_zero_values)
+
+    # Convert the list to a numpy array
+    non_zero_intensities_array = np.array(non_zero_intensities)
+
+    # Convert the list of intensities to a NumPy array and store in the dictionary
+    # Duplicate the array to have one array with 2 classes
+    processed_images_intensities = {}
+    processed_images_intensities['fat'] = np.array(non_zero_intensities_array)
+    processed_images_intensities['muscle'] = np.array(non_zero_intensities_array)
+
+    # Return the intensities arrays of the detected classes objects
+    return processed_images_intensities