Merge branch 'main' into AURORA-only

Author: evamariie

AURORA/tutorial.ipynb

@@ -1,6 +1,35 @@
 import os
 import matplotlib.pyplot as plt
 import nibabel as nib
+from pathlib import Path
+
+
+def visualize_data2(t1c, t1, t2, flair, height_p=0.5):  #
+    """Visualize the MRI modalities. This function is different from "visualize_data" in the following ways:
+        a) it expects the four file paths directly (no naming convention required)
+        b) it does not take a slice index but a percentage height (e.g. 0.5 = 50% = middle slice) as imputs may have different dimensions
+
+    Args:
+        height_p (float, optional): the percentage of slicing height. 0.5 means 50% means exactly in the middle of the cuboid
+        slice_index (int, optional): Slice to be visualized (first index in data of shape (155, 240, 240)). Defaults to 75.
+    """
+    _, axes = plt.subplots(1, 4, figsize=(12, 10))
+
+    for i, (mod, modality_file) in enumerate(
+        [("t1", t1), ("t1c", t1c), ("t2", t2), ("flair", flair)]
+    ):
+
+        data = nib.load(modality_file).get_fdata()
+        # Get the middle slice along the specified axis
+        slice_index = int(
+            data.shape[2] * height_p
+        )  # show slice that is exactly in the middle
+        slice_data = data[:, ::-1, slice_index].T
+
+        axes[i].set_title(mod)
+        axes[i].imshow(slice_data, cmap="gray")
+        axes[i].axis("off")
+
 
 DATA_FOLDER = "data"
 
@@ -23,7 +52,7 @@ def visualize_data(data_folder: str = DATA_FOLDER, slice_index: int = 75):
         axes[i].axis("off")
 
 
-def visualize_segmentation(modality_file: str, segmentation_file: str):
+def visualize_segmentation(modality_file: str, segmentation_file: str, slice_p=0.5):
     """Visualize the MRI modality and the segmentation
 
     Args:
@@ -34,10 +63,10 @@ def visualize_segmentation(modality_file: str, segmentation_file: str):
     seg_np = nib.load(segmentation_file).get_fdata().transpose(2, 1, 0)
     _, ax = plt.subplots(1, 2, figsize=(8, 4))
 
-    slice_index = modality_np.shape[0] // 2  # You can choose any slice here
+    slice_index = int(modality_np.shape[0] * slice_p)  # You can choose any slice here
     ax[0].imshow(modality_np[slice_index, :, :], cmap="gray")
     ax[1].imshow(modality_np[slice_index, :, :], cmap="gray")
-    ax[1].imshow(seg_np[slice_index, :, :], cmap="plasma", alpha=0.3)
+    ax[1].imshow(seg_np[slice_index, :, :], cmap="plasma", alpha=0.5)
     for ax in ax:
         ax.axis("off")
     plt.tight_layout()

BraTS-Toolkit/Pipeline_overview.jpg

@@ -3,9 +3,11 @@
 This folder contains several Jupyter notebooks to showcase different possible use cases of the [panoptica package](https://github.com/BrainLesion/panoptica).
 The package allows to compute instance-wise segmentation quality metrics for 2D and 3D semantic- and instance segmentation maps by providing 3 core modules:
 
-1. Instance Approximator: instance approximation algorithms in panoptic segmentation evaluation. Available now: connected components algorithm.
-1. Instance Matcher: instance matching algorithm in panoptic segmentation evaluation, to align and compare predicted instances with reference instances.
-1. Instance Evaluator: Evaluation of panoptic segmentation performance by evaluating matched instance pairs and calculating various metrics like true positives, Dice score, IoU, and ASSD for each instance.
+**1. Instance Approximator:** instance approximation algorithms in panoptic segmentation evaluation. Available now: connected components algorithm.
+
+**2. Instance Matcher:** instance matching algorithm in panoptic segmentation evaluation, to align and compare predicted instances with reference instances.
+
+**3. Instance Evaluator:** Evaluation of panoptic segmentation performance by evaluating matched instance pairs and calculating various metrics like true positives, Dice score, IoU, and ASSD for each instance.
 
 ![workflow_figure](https://github.com/BrainLesion/panoptica/blob/main/examples/figures/workflow.png)
 
@@ -17,7 +19,7 @@ The package allows to compute instance-wise segmentation quality metrics for 2D
 
 Although for many biomedical segmentation problems, an instance-wise evaluation is highly relevant and desirable, they are still addressed as semantic segmentation problems due to lack of appropriate instance labels.
 
-Modules [1-3] can be used to obtain panoptic metrics of matched instances based on a semantic segmentation input.
+**Modules [1-3]** can be used to obtain panoptic metrics of matched instances based on a semantic segmentation input.
 
 [Jupyter Notebook Example](https://github.com/BrainLesion/tutorials/tree/main/panoptica/example_spine_semantic.ipynb)
 
@@ -27,7 +29,7 @@ Modules [1-3] can be used to obtain panoptic metrics of matched instances based
 
 It is a common issue that instance segementation outputs have good segmentations with mismatched labels.
 
-For this case modules [2-3] can be utilized to match the instances and report panoptic metrics.
+For this case **modules [2-3]** can be utilized to match the instances and report panoptic metrics.
 
 [Jupyter Notebook Example](https://github.com/BrainLesion/tutorials/tree/main/panoptica/example_spine_unmatched_instance.ipynb)
 
@@ -37,7 +39,7 @@ For this case modules [2-3] can be utilized to match the instances and report pa
 
 Ideally the input data already provides matched instances.
 
-In this case module 3 can be used to directly report panoptic metrics without requiring any internal preprocessing.
+In this case **module 3** can be used to directly report panoptic metrics without requiring any internal preprocessing.
 
 [Jupyter Notebook Example](https://github.com/BrainLesion/tutorials/tree/main/panoptica/example_spine_matched_instance.ipynb)