Add resizing back

Seems to help for certain segmentation tasks.

Author: muellerflorian

debug.png

@@ -33,6 +33,20 @@ Results will be saved in the specified folder. For each image the following file
 
 ![segmentation__nuclei](img/segmentation__nuclei.png)
 
+### Resizing can speed up prediction & yield better results
+
+We found that resizing images before segmentation can yield better results for certain images. 
+Also, it will lead to faster computation time. In case you resize the images, we implemented a post-processing
+routine that will resize the predicted masks back to the original image size.
+
+Resizing can be specified in two ways
+
+1. **Scaling factor** recommended]: simply add an integer value. The actual image size will then be divided by this factor
+   to obtain the new size, i.e. a value of 2 will resize an image 512x512 to 256x256.
+2. **New size**: you can directly define the new size of the image, e.g. 256x256. Please note that this size
+   will be applied to all images, independly of their size. This option is hence not suitable if your data-sets 
+   contain differently sized images.
+
 ## Recommended workflow
 
 The default settings of the plugins allow to quickly perform the recommended workflow. You only have 
@@ -59,10 +73,12 @@ to paste your data folder.
     `String channel`    | str  |  `dapi` | Unique identifier to identify channel.
     `Object name`    | str  |  `nuclei` | How the object is called.
     `Cellpose model`    | str  |  `nuclei` | Cellpose model for segmentation: `cyto` or `nuclei`. Note that for dense nuclei, the cytoplasmic model might work better. 
+    `New size` | int | 2 | Numbers to specify resizing (see above). No resizing if empty.
     `Object diameter`     | int  | 50 | Typical diameter of the object. Better to be set a bit to small.
     `Net Average`     | Bool  | False | Can improve segmentation accuracy, but is slower (Runs the 4 built-in networks and averages them).
     `Resample`     | Bool  | False | Gives more accurate boundaries, but can be very slow (Runs dynamics at original image size).
     `String img ext`     | str  | `.png` | File extension of images that should be segmented.
+    `New size` | int | 2 | Numbers to specify resizing (see abov. No resizing if empty.
 
 2. Pressing on the plugin name `SegmentObjects` will start the segmentation.
     When using CellPose for the first time, the models for nuclear and cytoplasmic segmentations are downloaded. 
@@ -91,6 +107,7 @@ to paste your data folder.
     `String CELLS`    | str  |  `cy3` | Unique identifier for images of cytoplasmic stain.
     `String NUCLEI`    | str  |  `dapi` | Unique identifier for images of nuclear stain.
     `String img ext`     | str  | `.png` | File extension of images that should be segmented.
+    `New size` | int | 2 | Numbers to specify resizing (see above). No resizing if empty.
     `Size CELLS`     | int  | 100 | Typical size of a cell (in resized image).
     `Size NUCLEI`     | int  | 50 | Typical size of a nucleus (in resized image).
     `Net Average`     | Bool  | False | Can improve segmentation accuracy, but is slower (Runs the 4 built-in networks and averages them).

docs/analysis-segmentation.md

@@ -27,7 +27,7 @@ conda activate fq-segmentation
 **Install code Jupyter optimized for ImJoy**:
 
 ``` bash
-pip install -U imjoy
+pip install -U imjoy[jupyter]
 ```
 
 ## Connect to Jupyter engine

docs/img/imjoy-segment-cells-nuclei-ui.png

@@ -1,16 +1,13 @@
 <docs lang="markdown">
-ImJoy plugin to perform deep-learning based cell segmentation.
-
-- For more information, see the detailed documentation [**here**](https://fq-segmentation.readthedocs.io/en/latest/).
-- Used the the generalist cell/nuclei segmentation package [**Cellpose**](https://github.com/mouseland/cellpose). 
-
+ImJoy plugin to perform cell segmentation with 
+the generalist cell/nuclei segmentation package [**Cellpose**](https://github.com/mouseland/cellpose). 
 </docs>
 
 <config lang="json">
 {
   "name": "SegmentCellsNuclei",
   "type": "native-python",
-  "version": "0.2.0",
+  "version": "0.2.1",
   "description": "Segment cells AND nuclei.",
   "tags":  ["stable","dev"],
   "ui": [
@@ -19,6 +16,7 @@
         "<span style='font-weight: 550; color:#666362'>Path SAVE</span>: {id: 'path_save', type: 'string', placeholder: 'segmentation-input˃˃segmentation-results'}",
         "<span style='font-weight: 550; color:#666362'>String CELLS</span>: {id: 'str_cyto', type: 'string', placeholder: 'cy3'}",
         "<span style='font-weight: 550; color:#666362'>String NUCLEI</span>: {id: 'str_nuclei', type: 'string', placeholder: 'dapi'}",
+        "<span style='font-weight: 550; color:#666362'>Scaling factor/New size: </span>: {id: 'new_size', type: 'string', placeholder: '2'}",
         "<span style='font-weight: 550; color:#666362'>Size CELLS</span>: {id: 'size_cells', type: 'number', placeholder: 200}",
         "<span style='font-weight: 550; color:#666362'>Size NUCLEI</span>: {id: 'size_nuclei', type: 'number', placeholder: 100}",
         "<span style='font-weight: 550; color:#666362'>Net Average</span>: {id: 'net_avg', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
@@ -38,7 +36,7 @@
                     "stable":["pip: -U git+https://github.com/fish-quant/fq-segmentation@master"]
                    },
   "dependencies": [],
-  "cover": "https://raw.githubusercontent.com/fish-quant/fq-segmentation/master/docs/img/segmentation__cells.png"
+  "cover": "https://fish-quant.github.io/fq-segmentation/img/segmentation__cells.png"
 }
 </config>
 
@@ -114,13 +112,32 @@
         else:
             net_avg = False
 
+        # >> Check new size
+        new_size_str = ctx.config.new_size
+
+        if len(new_size_str) == 0: 
+            new_size = ()
+
+        else:
+            if new_size_str.count(',') == 0:
+                new_size = (int(new_size_str),)
+
+            elif new_size_str.count(',') == 1:
+                new_size = tuple(int(x) for x in new_size_str.split(','))
+
+            else:
+                api.alert('String to define new size has to contain one or zero \",\" to separate the new dimensions in XY.')
+                return
+
+        api.log(f'New size: {new_size}')
 
         # >>>> Call pre-processing function
         api.showStatus('Performing segmentation ... see log for details')
         utils_cellpose.segment_cells_nuclei_indiv(
                                       path_scan=path_scan, 
                                       str_channels=(ctx.config.str_cyto, ctx.config.str_nuclei), 
                                       img_ext=ctx.config.img_ext, 
+                                      new_size=new_size,
                                       model_types=('cyto','nuclei'),
                                       diameters=(ctx.config.size_cells, ctx.config.size_nuclei), 
                                       net_avg=net_avg, 

docs/img/imjoy-segment-objects-ui.png

@@ -1,16 +1,13 @@
 <docs lang="markdown">
-ImJoy plugin to perform deep-learning based cell segmentation.
-
-- For more information, see the detailed documentation [**here**](https://fq-segmentation.readthedocs.io/en/latest/).
-- Used the the generalist cell/nuclei segmentation package [**Cellpose**](https://github.com/mouseland/cellpose). 
-
+ImJoy plugin to perform object segmentation with 
+the generalist cell/nuclei segmentation package [**Cellpose**](https://github.com/mouseland/cellpose). 
 </docs>
 
 <config lang="json">
 {
   "name": "SegmentObjects",
   "type": "native-python",
-  "version": "0.2.0",
+  "version": "0.2.1",
   "description": "Segment cells OR nuclei.",
   "tags":  ["stable","dev"],
   "ui": [
@@ -20,6 +17,7 @@
         "<span style='font-weight: 550; color:#666362'>String channel</span>: {id: 'str_channel', type: 'string', placeholder: 'dapi'}",
         "<span style='font-weight: 550; color:#666362'>Object name</span>: {id: 'obj_name', type: 'string', placeholder: 'nuclei'}",
         "<span style='font-weight: 550; color:#666362'>Cellpose model</span>: {id: 'model_type', type: 'choose', options: ['nuclei', 'cyto'], placeholder: 'nuclei'}",
+        "<span style='font-weight: 550; color:#666362'>Scaling factor/New size:</span>: {id: 'new_size', type: 'string', placeholder: '1'}",
         "<span style='font-weight: 550; color:#666362'>Object diameter</span>: {id: 'diameter', type: 'number', placeholder: 100}",
         "<span style='font-weight: 550; color:#666362'>Net Average</span>: {id: 'net_avg', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
         "<span style='font-weight: 550; color:#666362'>Resample</span>: {id: 'resample', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
@@ -38,7 +36,7 @@
                     "stable":["pip: -U git+https://github.com/fish-quant/fq-segmentation@master"]
                    },
   "dependencies": [],
-  "cover": "https://raw.githubusercontent.com/fish-quant/fq-segmentation/master/docs/img/segmentation__nuclei.png"
+  "cover": "https://fish-quant.github.io/fq-segmentation/img/segmentation__cells.png"
 }
 </config>
 
@@ -113,6 +111,24 @@
         else:
             net_avg = False
 
+        # >> Check new size
+        new_size_str = ctx.config.new_size
+
+        if len(new_size_str) == 0: 
+            new_size = ()
+
+        else:
+            if new_size_str.count(',') == 0:
+                new_size = (int(new_size_str),)
+
+            elif new_size_str.count(',') == 1:
+                new_size = tuple(int(x) for x in new_size_str.split(','))
+
+            else:
+                api.alert('String to define new size has to contain one or zero \",\" to separate the new dimensions in XY.')
+                return
+
+        api.log(f'New size: {new_size}')
 
         # >>> Call segmentation function
         api.showStatus('Performing segmentation ... see log for details')
@@ -122,6 +138,7 @@
                                 obj_name=ctx.config.obj_name,
                                 str_channel=ctx.config.str_channel, 
                                 img_ext=ctx.config.img_ext, 
+                                new_size=new_size,
                                 model_type=ctx.config.model_type,
                                 diameter= ctx.config.diameter, 
                                 net_avg=net_avg, 

docs/imjoy-installation.md

@@ -81,7 +81,7 @@
   {
    "name": "SegmentCellsNuclei",
    "type": "native-python",
-   "version": "0.2.0",
+   "version": "0.2.1",
    "description": "Segment cells AND nuclei.",
    "tags": [
     "stable",
@@ -93,13 +93,14 @@
     "<span style='font-weight: 550; color:#666362'>Path SAVE</span>: {id: 'path_save', type: 'string', placeholder: 'segmentation-input˃˃segmentation-results'}",
     "<span style='font-weight: 550; color:#666362'>String CELLS</span>: {id: 'str_cyto', type: 'string', placeholder: 'cy3'}",
     "<span style='font-weight: 550; color:#666362'>String NUCLEI</span>: {id: 'str_nuclei', type: 'string', placeholder: 'dapi'}",
+    "<span style='font-weight: 550; color:#666362'>Scaling factor/New size: </span>: {id: 'new_size', type: 'string', placeholder: '2'}",
     "<span style='font-weight: 550; color:#666362'>Size CELLS</span>: {id: 'size_cells', type: 'number', placeholder: 200}",
     "<span style='font-weight: 550; color:#666362'>Size NUCLEI</span>: {id: 'size_nuclei', type: 'number', placeholder: 100}",
     "<span style='font-weight: 550; color:#666362'>Net Average</span>: {id: 'net_avg', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
     "<span style='font-weight: 550; color:#666362'>Resample</span>: {id: 'resample', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
     "<span style='font-weight: 550; color:#666362'>String img ext</span>: {id: 'img_ext', type: 'string', placeholder: '.png'}"
    ],
-   "cover": "https://raw.githubusercontent.com/fish-quant/fq-segmentation/master/docs/img/segmentation__cells.png",
+   "cover": "https://fish-quant.github.io/fq-segmentation/img/segmentation__cells.png",
    "inputs": null,
    "outputs": null,
    "flags": [],
@@ -121,7 +122,7 @@
   {
    "name": "SegmentObjects",
    "type": "native-python",
-   "version": "0.2.0",
+   "version": "0.2.1",
    "description": "Segment cells OR nuclei.",
    "tags": [
     "stable",
@@ -134,12 +135,13 @@
     "<span style='font-weight: 550; color:#666362'>String channel</span>: {id: 'str_channel', type: 'string', placeholder: 'dapi'}",
     "<span style='font-weight: 550; color:#666362'>Object name</span>: {id: 'obj_name', type: 'string', placeholder: 'nuclei'}",
     "<span style='font-weight: 550; color:#666362'>Cellpose model</span>: {id: 'model_type', type: 'choose', options: ['nuclei', 'cyto'], placeholder: 'nuclei'}",
+    "<span style='font-weight: 550; color:#666362'>Scaling factor/New size:</span>: {id: 'new_size', type: 'string', placeholder: '1'}",
     "<span style='font-weight: 550; color:#666362'>Object diameter</span>: {id: 'diameter', type: 'number', placeholder: 100}",
     "<span style='font-weight: 550; color:#666362'>Net Average</span>: {id: 'net_avg', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
     "<span style='font-weight: 550; color:#666362'>Resample</span>: {id: 'resample', type: 'choose', options: ['false', 'true'], placeholder: 'false'}",
     "<span style='font-weight: 550; color:#666362'>String img ext</span>: {id: 'img_ext', type: 'string', placeholder: '.png'}"
    ],
-   "cover": "https://raw.githubusercontent.com/fish-quant/fq-segmentation/master/docs/img/segmentation__nuclei.png",
+   "cover": "https://fish-quant.github.io/fq-segmentation/img/segmentation__cells.png",
    "inputs": null,
    "outputs": null,
    "flags": [],

imjoy-plugins/SegmentCellsNuclei.imjoy.html

@@ -42,8 +42,8 @@
     "str_cyto = 'cy3'                                   # Identifier of channel for cytoplasmic segmentation\n",
     "str_nuclei = 'dapi'                                # Identifier of channel for nuclear segmentation\n",
     "img_ext = '.png'                                   # Extension of images to be segmented\n",
-    "new_size = (2,)                                    # Size of images (when resize should be applied, empty tuple otherwise)\n",
     "\n",
+    "new_size = (2,)                                    # Size of images (when resize should be applied, empty tuple otherwise)\n",
     "diameter_cells = 100                               # Typical size (diameter) of cells\n",
     "diameter_nuclei = 50                               # Typical size (diameter) of nuclei\n",
     "\n",
@@ -55,6 +55,7 @@
     "                                path_scan=path_scan, \n",
     "                                str_channels=str_channels, \n",
     "                                img_ext=img_ext,  \n",
+    "                                new_size=new_size,\n",
     "                                model_types=('cyto','nuclei'), \n",
     "                                diameters = (diameter_cells, diameter_nuclei), \n",
     "                                net_avg=net_avg, \n",

imjoy-plugins/SegmentObjects.imjoy.html

@@ -42,7 +42,10 @@
     "obj_name='nuclei',                                 # Name of object that should be segmented\n",
     "str_channel = 'dapi'                                # Identifier of channel for nuclear segmentation\n",
     "img_ext = '.png'                                   # Extension of images to be segmented\n",
+    "\n",
+    "new_size = (2,)                                    # Size of images (when resize should be applied, empty tuple otherwise)\n",
     "diameter = 50                                      # Typical size (diameter) of nuclei\n",
+    "\n",
     "model_type = 'nuclei'                              # Can be 'nuclei' or 'cyto', for densely packed nuclei 'cyto' might work well. \n",
     "net_avg=False                                      #  runs the 4 built-in networks and averages them if True, runs one network if False\n",
     "resample=False                                     #  run dynamics at original image size (will be (much) slower but create more accurate boundaries)\n",
@@ -52,7 +55,8 @@
     "                        path_scan=path_scan, \n",
     "                        obj_name=obj_name, \n",
     "                        str_channel=str_channel, \n",
-    "                        img_ext=img_ext,  \n",
+    "                        img_ext=img_ext,\n",
+    "                        new_size=new_size,  \n",
     "                        model_type=model_type, \n",
     "                        diameter = diameter, \n",
     "                        net_avg=net_avg, \n",

manifest.imjoy.json

@@ -12,4 +12,4 @@
 # MINOR: new features
 # PATCH: backwards compatible bug fixes
 # MAJOR.MINOR.PATCHdev means a version under development
-__version__ = "0.3.1"
+__version__ = "0.3.2"