feat: add N4BiasFieldCorrection transform class

src/imgtools/transforms/intensity_transforms.py

@@ -1,12 +1,14 @@
 from dataclasses import dataclass
 
 from SimpleITK import Image
+from SimpleITK import N4BiasFieldCorrectionImageFilter
 
 from .base_transform import BaseTransform
 from .functional import (
     clip_intensity,
     window_intensity,
 )
+from .lambda_transforms import SimpleITKFilter
 
 __all__ = [
     "IntensityTransform",
@@ -150,3 +152,29 @@ def __call__(self, image: Image) -> Image:
         """
 
         return window_intensity(image, self.window, self.level)
+
+@dataclass
+class N4BiasFieldCorrection(IntensityTransform):
+    """
+    N4 bias-field correction.
+    - Corrects smooth intensity inhomogeneity (bias field) in the image.
+    - Does NOT change image geometry (no rotation, translation, or scaling).
+    """
+
+    def __post_init__(self) -> None:
+        self.filter = SimpleITKFilter(N4BiasFieldCorrectionImageFilter())
+
+    def __call__(self, image: Image) -> Image:
+        """Apply N4 bias-field intensity correction to an image.
+
+        Parameters
+        ----------
+        image : Image
+            The input intensity image (e.g., MRI).
+        Returns
+        -------
+        Image
+            Image with corrected intensities and unchanged geometry.
+        """
+        return self.filter(image)
+

src/imgtools/transforms/transformer.py

@@ -11,6 +11,7 @@
     IntensityTransform,
     SpatialTransform,
 )
+from imgtools.transforms.intensity_transforms import N4BiasFieldCorrection
 
 # Define TypeVars for the different image types
 T_MedImage = TypeVar("T_MedImage", bound=MedImage)
@@ -68,6 +69,11 @@ def _apply_transforms(
                     and transform.supports_reference()
                 ):
                     transformed_image = transform(transformed_image, ref)
+                elif (
+                    isinstance(transform, N4BiasFieldCorrection) 
+                    and transformed_image.metadata.get("Modality", "Unknown") == "MR"
+                ):
+                    transformed_image = transform(transformed_image)
                 elif isinstance(
                     transform, (IntensityTransform, SpatialTransform)
                 ):