Add more info in the new synapse detection pipeline

Author: danifranco

biapy/config/config.py

@@ -72,11 +72,14 @@ def __init__(self, job_dir: str, job_identifier: str):
         #   - 'F' stands for 'Flow' where, for each instance, contains the distance values to its center of mass for each dimension.
         #
         #### For "synapse" type of instances ####
-        # Possible options: 'BF'. This variable defines the channels to be used to represent synapse instances based on the input 
+        # Possible options: 'B' and 'BF' (still experimental). This variable defines the channels to be used to represent synapse instances based on the input 
         # synapse sites. The meaning of each letter is a follows:
-        #   - 'B' stands for 'Binary mask', it is a binary representation of each postsynaptic site
-        #   - 'F' stands for 'Flow' and contains the distance values to the corresponding presynaptic site (of each postsynaptic 
-        #     site) for each dimension.
+        #   * In 'B', the first and second channels represent the pre and post points as 3D points, respectively.
+        #
+        #   * In 'BF':
+        #       - 'B' stands for 'Binary mask', it is a binary representation of each postsynaptic site
+        #       - 'F' stands for 'Flow' and contains the distance values to the corresponding presynaptic site (of each postsynaptic 
+        #         site) for each dimension.¡
         _C.PROBLEM.INSTANCE_SEG.DATA_CHANNELS = "BC"
         # Whether to mask the distance channel to only calculate the loss in those regions where the binary mask
         # defined by B channel is present
@@ -145,10 +148,10 @@ def __init__(self, job_dir: str, job_identifier: str):
         # Options available: 
         #   * 'auto' to decide the threshold to be applied by measuring it with Otsu
         #   * 'manual' to set a fixed threshold defined by 'PROBLEM.INSTANCE_SEG.SYNAPSES.MIN_TH_TO_BE_PEAK'
-        #   * 'relative_by_patch' to use 'PROBLEM.INSTANCE_SEG.SYNAPSES.MIN_TH_TO_BE_PEAK' but relative to the maximum value in the predicted data. 
-        #     More info in https://scikit-image.org/docs/0.25.x/api/skimage.feature.html#skimage.feature.peak_local_max and 
+        #   * 'relative_by_patch' to use 'PROBLEM.INSTANCE_SEG.SYNAPSES.MIN_TH_TO_BE_PEAK' but relative to the maximum value in the predicted patch 
+        #     data. More info in https://scikit-image.org/docs/0.25.x/api/skimage.feature.html#skimage.feature.peak_local_max and 
         #     https://scikit-image.org/docs/0.23.x/api/skimage.feature.html#skimage.feature.blob_log (see 'threshold_rel' argument description)
-        #   * 'relative' to use 'PROBLEM.INSTANCE_SEG.SYNAPSES.MIN_TH_TO_BE_PEAK' but relative to the maximum value in the predicted data. 
+        #   * 'relative' to use 'PROBLEM.INSTANCE_SEG.SYNAPSES.MIN_TH_TO_BE_PEAK' but relative to the maximum value in the whole predicted data. 
         #     More info in https://scikit-image.org/docs/0.25.x/api/skimage.feature.html#skimage.feature.peak_local_max and 
         #     https://scikit-image.org/docs/0.23.x/api/skimage.feature.html#skimage.feature.blob_log (see 'threshold_rel' argument description)
         _C.PROBLEM.INSTANCE_SEG.SYNAPSES.TH_TYPE = "auto"

biapy/data/post_processing/post_processing.py

@@ -391,7 +391,8 @@ def erode_seed_and_foreground():
             )
     return segm
 
-def create_synapses(data: NDArray,
+def create_synapses(
+    data: NDArray,
     channels: str,
     point_creation_func: str = "peak_local_max",
     min_th_to_be_peak: List[float] = [0.2],
@@ -403,19 +404,49 @@ def create_synapses(data: NDArray,
     relative_th_value: bool = False,
 ) -> Tuple[NDArray, Dict]:
     """
-    Convert binary foreground probability maps and instance contours to instance masks via watershed segmentation
-    algorithm.
-
-    Implementation based on `PyTorch Connectomics' process.py
-    <https://github.com/zudi-lin/pytorch_connectomics/blob/master/connectomics/utils/process.py>`_.
+    Creates synapses pre/post points from the given ``data``. 
 
+    Find more info regarding ``min_distance``, ``min_sigma`` and ``exclude_border`` arguments in 
+    `peak_local_max <https://scikit-image.org/docs/0.25.x/api/skimage.feature.html#skimage.feature.peak_local_max>`__ 
+    and `blob_log <https://scikit-image.org/docs/0.25.x/api/skimage.feature.html#skimage.feature.blob_log>`__ 
+    functions.  
+    
     Parameters
     ----------
     data : 4D Numpy array
         Binary foreground labels and contours data to apply watershed into. E.g. ``(397, 1450, 2000, 2)``.
 
     channels : str
         Channel type used. Possible options: ``A``, ``C``, ``BC``, ``BCM``, ``BCD``, ``BCDv2``, ``Dv2`` and ``BDv2``.
+
+    point_creation_func : str, optional
+        Function to be used in the pre/post point creation. Options: ["peak_local_max", "blob_log"]
+    
+    min_th_to_be_peak : List of float, optional
+        Minimum values for each channel in ``data`` to be used as the minimum when creating the points. 
+        
+    min_distance : int, optional
+        The minimal allowed distance separating peaks. To find the maximum number of peaks, use ``min_distance=1``.
+
+    min_sigma : int, optional
+        The minimum standard deviation for Gaussian kernel. Keep this low to detect smaller blobs. The standard deviations
+        of the Gaussian filter are given for each axis as a sequence, or as a single number, in which case it is equal for
+        all axes.
+        
+    max_sigma : int, optional
+        The maximum standard deviation for Gaussian kernel. Keep this high to detect larger blobs. The standard deviations 
+        of the Gaussian filter are given for each axis as a sequence, or as a single number, in which case it is equal for 
+        all axes.
+
+    num_sigma : int, optional
+        The number of intermediate values of standard deviations to consider between ``min_sigma`` and ``max_sigma``.
+
+    exclude_border: bool, optional
+        To exclude border-pixels of the border of the image.
+
+    relative_th_value : bool, optional
+        To use ``min_th_to_be_peak`` as a relative threshold to the maximum value, i.e. ``threshold_rel`` is set instead of 
+        ``threshold_abs`` in the ``peak_local_max`` or ``blob_log`` call. 
     """
     assert channels in [
         "B",