Merge pull request #87 from qupath/0.4

Add WSInfer TIL script

docs/advanced/exporting_annotations.md

@@ -1,3 +1,4 @@
+(exporting-annotations)=
 # Exporting annotations
 
 People often ask about exporting QuPath annotations, to which the response is invariably *'In which format exactly?'*

docs/advanced/exporting_images.md

@@ -31,6 +31,7 @@ Exporting an RGB image as a JPEG, for example, may introduce new compression art
 The point is that no bit-depth or channel conversion is involved.
 :::
 
+(exporting-rendered-images)=
 ### Rendered images
 
 Images within QuPath can have different numbers of channels and various bit-depths, but when they are displayed they generally need to be *rendered* as [3-channel, 8-bit RGB](https://petebankhead.gitbooks.io/imagej-intro/content/chapters/colors/colors.html#rgb-images).

docs/deep/wsinfer.md

@@ -3,7 +3,7 @@
 
 The [WSInfer QuPath extension](https://github.com/qupath/qupath-extension-wsinfer/) makes it possible to do patch-based deep learning inference for digital pathology, without any need for scripting.
 
-It's a collaboration between the QuPath group (the extension) and Stony Brook University ([WSInfer](https://wsinfer.readthedocs.io/en/latest/)).
+It's a collaboration between Stony Brook University ([WSInfer](https://wsinfer.readthedocs.io/en/latest/)) and the QuPath group (the [extension](https://github.com/qupath/qupath-extension-wsinfer/) that brings WSInfer models to QuPath).
 
 :::{admonition} Cite the paper!
 :class: warning
@@ -17,6 +17,12 @@ If you use WSInfer and/or this extension in a publication, please make sure to c
 - [WSInfer QuPath Extension](https://github.com/qupath/qupath-extension-wsinfer/releases)
 - PyTorch (this can be downloaded while using the extension)
 
+:::{tip}
+A GPU is not required but can dramatically speed up processing.
+If you have an NVIDIA GPU and want to use it with WSInfer, you will need to install a version of CUDA compatible with PyTorch - please see [the Deep Java Library page](deep-java-library-gpu).
+:::
+
+
 ## Set-up
 
 With QuPath installed and running, drag and drop the WSInfer extension into the application and restart QuPath.
@@ -129,7 +135,7 @@ selectAnnotations()
 qupath.ext.wsinfer.WSInfer.runInference("kaczmarj/pancancer-lymphocytes-inceptionv4.tcga")
 ```
 
-where the `selectAnnotation()` line was added when I pressed the {guilabel}`Annotations` button in the WSInfer dialog, and the following line runs the specified models (creating tiles automatically).
+where the `selectAnnotation()` line was added when I pressed the {guilabel}`Annotations` button in the WSInfer dialog, and the following line runs the specified model (creating tiles automatically).
 
 To process in batch, I would need to
 
@@ -138,3 +144,15 @@ To process in batch, I would need to
 * Open the above script in QuPath's script editor
 * Choose {menuselection}`Run --> Run for project`, and select the images I want to process
 
+
+## Identifying TILs (overlaying predictions of two models)
+
+By combining multiple WSInfer models, it's possible to develop more complex end-to-end workflows in QuPath.
+This is easiest to achieve using scripting.
+
+The script below aims to identify regions containing **tumor-infiltrating lymphocytes (TILs)**.
+It does this by applying a lymphocyte patch classification model and then a breast tumor classification model on the same patches.
+
+After running the script, you can [export results](exporting-measurements) or [markup images](exporting-rendered-images).
+
+<script src="https://gist.github.com/petebankhead/aebd135d3f5a080f6216fb05d8029c42.js"></script>

docs/tutorials/exporting_measurements.md

@@ -1,3 +1,4 @@
+(exporting-measurements)=
 # Exporting measurements
 
 There are 3 different ways to export measurements within QuPath, via: