Merge pull request #296 from nanli-emory/num_tissues

get num tissues
choosehappy · Jul 22, 2024 · 26a98cb · 26a98cb
2 parents 6f0bacd + 222c372
commit 26a98cb
Show file tree

Hide file tree

Showing 5 changed files with 97 additions and 55 deletions.
diff --git a/histoqc/BaseImage.py b/histoqc/BaseImage.py
@@ -8,7 +8,7 @@
 from typing import Union, Tuple
 #os.environ['PATH'] = 'C:\\research\\openslide\\bin' + ';' + os.environ['PATH'] #can either specify openslide bin path in PATH, or add it dynamically
 from histoqc.import_wrapper.openslide import openslide
-
+from skimage import measure, morphology
 # there is no branch reset group in re
 # compatible with the previous definition of valid input: leading zero and leading decimals are supported
 _REGEX_MAG = r"^(\d?\.?\d*X?)"
@@ -206,6 +206,9 @@ def getImgThumb(self, size: str):
                 self[key] = getBestThumb(self, bx, by, target_dims, target_sampling_factor)
         return self[key]
 
+'''
+the following are helper functions 
+'''
 def getBestThumb(s: BaseImage, x: int, y: int, dims: Tuple[int, int], target_sampling_factor: float):
     osh = s["os_handle"]
 
@@ -224,9 +227,7 @@ def getBestThumb(s: BaseImage, x: int, y: int, dims: Tuple[int, int], target_sam
     else:
         return resizeTileDownward(s, target_sampling_factor, level)
 
-'''
-the followings are helper functions 
-'''
+
 def resizeTileDownward(self, target_downsampling_factor, level):
     osh = self["os_handle"]
     (bx, by, bwidth, bheight) = self["img_bbox"]
@@ -341,3 +342,22 @@ def getDimensionsByOneDim(s: BaseImage, dim: int) -> Tuple[int, int]:
     else:
         w = int(dim * width / height)
         return w, dim
+
+
+def getMaskRegionsStats(img)-> dict:
+
+    rps = measure.regionprops(morphology.label(img))
+    if rps:
+        areas = np.asarray([rp.area for rp in rps])
+        num = len(rps)
+        area_min = areas.min()
+        area_max = areas.max()
+        area_mean = areas.mean()
+    else:
+        num = area_min = area_max = area_mean = 0
+    return {
+        'num': num,
+        'area_min': area_min,
+        'area_max': area_max,
+        'area_mean': area_mean
+    }
diff --git a/histoqc/BasicModule.py b/histoqc/BasicModule.py
@@ -1,6 +1,6 @@
 import logging
 import os
-from histoqc.BaseImage import printMaskHelper
+from histoqc.BaseImage import printMaskHelper, getMaskRegionsStats
 from skimage.morphology import remove_small_objects, binary_opening, disk
 from skimage import io, color, img_as_ubyte
 
@@ -69,3 +69,9 @@ def finalProcessingArea(s, params):
             f"{s['filename']} - After BasicModule.finalProcessingArea NO tissue remains detectable! Downstream modules likely to be incorrect/fail")
         s["warnings"].append(
             f"After BasicModule.finalProcessingArea NO tissue remains detectable! Downstream modules likely to be incorrect/fail")
+
+
+def countTissuePieces(s):
+    mask = s["img_mask_use"]
+    stats = getMaskRegionsStats(mask)
+    s.addToPrintList("#pieces_of_tissue", str(stats.get('num', 0)))
diff --git a/histoqc/BlurDetectionModule.py b/histoqc/BlurDetectionModule.py
@@ -2,7 +2,7 @@
 import os
 
 import skimage
-from histoqc.BaseImage import printMaskHelper
+from histoqc.BaseImage import printMaskHelper, getMaskRegionsStats
 from skimage import io, img_as_ubyte, morphology, measure
 from skimage.color import rgb2gray
 from skimage.filters import rank
@@ -37,19 +37,20 @@ def identifyBlurryRegions(s, params):
     prev_mask = s["img_mask_use"]
     s["img_mask_use"] = s["img_mask_use"] & ~s["img_mask_blurry"]
 
-    rps = measure.regionprops(morphology.label(mask))
-    if rps:
-        areas = np.asarray([rp.area for rp in rps])
-        nobj = len(rps)
-        area_max = areas.max()
-        area_mean = areas.mean()
-    else:
-        nobj = area_max = area_mean = 0
-
-
-    s.addToPrintList("blurry_removed_num_regions", str(nobj))
-    s.addToPrintList("blurry_removed_mean_area", str(area_mean))
-    s.addToPrintList("blurry_removed_max_area", str(area_max))
+    # rps = measure.regionprops(morphology.label(mask))
+    # if rps:
+    #     areas = np.asarray([rp.area for rp in rps])
+    #     nobj = len(rps)
+    #     area_max = areas.max()
+    #     area_mean = areas.mean()
+    # else:
+    #     nobj = area_max = area_mean = 0
+    stat_rs = getMaskRegionsStats(mask)
+
+
+    s.addToPrintList("blurry_removed_num_regions", str(stat_rs.get('num', 0)))
+    s.addToPrintList("blurry_removed_mean_area", str(stat_rs.get('area_mean',0)))
+    s.addToPrintList("blurry_removed_max_area", str(stat_rs.get('area_max',0)))
 
 
     s.addToPrintList("blurry_removed_percent",

diff --git a/histoqc/MorphologyModule.py b/histoqc/MorphologyModule.py
@@ -1,7 +1,7 @@
 import logging
 import os
 import numpy as np
-from histoqc.BaseImage import printMaskHelper
+from histoqc.BaseImage import printMaskHelper, getMaskRegionsStats
 from skimage import io, morphology, img_as_ubyte, measure
 
 from scipy import ndimage as ndi
@@ -23,19 +23,23 @@ def removeSmallObjects(s, params):
     s["img_mask_use"] = img_reduced
 
 
-    rps = measure.regionprops(morphology.label(img_small))
-    if rps:
-        areas = np.asarray([rp.area for rp in rps])
-        nobj = len(rps)
-        area_max = areas.max()
-        area_mean = areas.mean()
-    else:
-        nobj = area_max = area_mean = 0
+    # rps = measure.regionprops(morphology.label(img_small))
+    # if rps:
+    #     areas = np.asarray([rp.area for rp in rps])
+    #     nobj = len(rps)
+    #     area_max = areas.max()
+    #     area_mean = areas.mean()
+    # else:
+    #     nobj = area_max = area_mean = 0
+    stats = getMaskRegionsStats(img_small)
 
-    s.addToPrintList("small_tissue_removed_num_regions", str(nobj))
-    s.addToPrintList("small_tissue_removed_mean_area", str(area_mean))
-    s.addToPrintList("small_tissue_removed_max_area", str(area_max))
 
+    # s.addToPrintList("small_tissue_removed_num_regions", str(nobj))
+    # s.addToPrintList("small_tissue_removed_mean_area", str(area_mean))
+    # s.addToPrintList("small_tissue_removed_max_area", str(area_max))
+    s.addToPrintList("small_tissue_removed_num_regions", str(stats.get('num',0)))
+    s.addToPrintList("small_tissue_removed_mean_area", str(stats.get('area_mean',0)))
+    s.addToPrintList("small_tissue_removed_max_area", str(stats.get('area_max',0)))
 
 
 
@@ -87,18 +91,22 @@ def removeFatlikeTissue(s, params):
     prev_mask = s["img_mask_use"]
     s["img_mask_use"] = prev_mask & ~mask_fat
 
-    rps = measure.regionprops(morphology.label(mask_fat))
-    if rps:
-        areas = np.asarray([rp.area for rp in rps])
-        nobj = len(rps)
-        area_max = areas.max()
-        area_mean = areas.mean()
-    else:
-        nobj = area_max = area_mean = 0
+    # rps = measure.regionprops(morphology.label(mask_fat))
+    # if rps:
+    #     areas = np.asarray([rp.area for rp in rps])
+    #     nobj = len(rps)
+    #     area_max = areas.max()
+    #     area_mean = areas.mean()
+    # else:
+    #     nobj = area_max = area_mean = 0
+    stats = getMaskRegionsStats(mask_fat)
 
-    s.addToPrintList("fatlike_tissue_removed_num_regions", str(nobj))
-    s.addToPrintList("fatlike_tissue_removed_mean_area", str(area_mean))
-    s.addToPrintList("fatlike_tissue_removed_max_area", str(area_max))
+    # s.addToPrintList("fatlike_tissue_removed_num_regions", str(nobj))
+    # s.addToPrintList("fatlike_tissue_removed_mean_area", str(area_mean))
+    # s.addToPrintList("fatlike_tissue_removed_max_area", str(area_max))
+    s.addToPrintList("fatlike_tissue_removed_num_regions", str(stats.get('num',0)))
+    s.addToPrintList("fatlike_tissue_removed_mean_area", str(stats.get('area_mean',0)))
+    s.addToPrintList("fatlike_tissue_removed_max_area", str(stats.get('area_max',0)))
 
 
 
@@ -125,18 +133,22 @@ def fillSmallHoles(s, params):
     prev_mask = s["img_mask_use"]
     s["img_mask_use"] = img_reduced
 
-    rps = measure.regionprops(morphology.label(img_small))
-    if rps:
-        areas = np.asarray([rp.area for rp in rps])
-        nobj = len(rps)
-        area_max = areas.max()
-        area_mean = areas.mean()
-    else:
-        nobj = area_max = area_mean = 0
-
-    s.addToPrintList("small_tissue_filled_num_regions", str(nobj))
-    s.addToPrintList("small_tissue_filled_mean_area", str(area_mean))
-    s.addToPrintList("small_tissue_filled_max_area", str(area_max))
+    # rps = measure.regionprops(morphology.label(img_small))
+    # if rps:
+    #     areas = np.asarray([rp.area for rp in rps])
+    #     nobj = len(rps)
+    #     area_max = areas.max()
+    #     area_mean = areas.mean()
+    # else:
+    #     nobj = area_max = area_mean = 0
+    stats = getMaskRegionsStats(img_small)
+
+    # s.addToPrintList("small_tissue_filled_num_regions", str(nobj))
+    # s.addToPrintList("small_tissue_filled_mean_area", str(area_mean))
+    # s.addToPrintList("small_tissue_filled_max_area", str(area_max))
+    s.addToPrintList("small_tissue_filled_num_regions", str(stats.get('num', 0)))
+    s.addToPrintList("small_tissue_filled_mean_area", str(stats.get('area_mean', 0)))
+    s.addToPrintList("small_tissue_filled_max_area", str(stats.get('area_max', 0)))
 
     s.addToPrintList("small_tissue_filled_percent",
                      printMaskHelper(params.get("mask_statistics", s["mask_statistics"]), prev_mask, s["img_mask_use"]))

diff --git a/histoqc/config/config_v2.1.ini b/histoqc/config/config_v2.1.ini
@@ -27,8 +27,9 @@ steps= BasicModule.getBasicStats
     SaveModule.saveFinalMask
     SaveModule.saveMacro
     SaveModule.saveThumbnails
+    BasicModule.countTissuePieces
     BasicModule.finalComputations
-
+    
 
 [BaseImage.BaseImage]
 image_work_size = 1.25x
@@ -252,3 +253,5 @@ mask_name: img_mask_use
 disk_size: 5
 upper_threshold: 210
 invert: True
+
+[BasicModule.countTissuePieces]