Getting error: Standard exception caught in Qt event handler invalid vector subscript and with another model of breast seg model I get FAST exception caught in Qt event handler input port 1 on process object Segmentation Network is missing its required connection

[lomshabhishek]

model = fast.DataHub().download('breast-tumour-segmentation-model')
importer = fast.WholeSlideImageImporter.create(
image_path)

tissueSegmentation = fast.TissueSegmentation.create()
.connect(importer)

generator = fast.PatchGenerator.create(
256, 256,
magnification=20,
overlapPercent=0
).connect(importer)
.connect(1, tissueSegmentation)

segmentation = fast.SegmentationNetwork.create(
model.paths[0] + '/unet_tumour_refinement_model_fix-opset9.onnx',#high_res_nuclei_unet.onnx',
scaleFactor=1./255.
).connect(generator)

stitcher = fast.PatchStitcher.create()
.connect(segmentation)

renderer = fast.ImagePyramidRenderer.create()
.connect(importer)

segmentationRenderer = fast.SegmentationRenderer.create()
.connect(stitcher)

fast.SimpleWindow2D.create()
.connect(renderer)
.connect(segmentationRenderer)
.run()

[smistad]

"input port 1 on process object Segmentation Network is missing its required connection" happens here because the unet_tumour_refinement_model_fix-opset9.onnx model is a refinement model which has 2 inputs. Also, this is not a patch-wise model.
You can read about these two models in André's paper: https://www.frontiersin.org/articles/10.3389/fmed.2022.971873/full
Figure 3 shows how these two models are used.

Not sure about the other error, but I can look into it.

[smistad]

The other error you have is because the patch wise breast cancer model "pw_tumour_mobilenetv2_model.onnx" is not a segmentation network it is a image classification network. Thus you have to run it through NeuralNetwork not SegmentationNetwork and then display the results using heatmap renderer. Here is an example, in which non-tumour areas are highlighted as green, and tumour areas are highlighted as red:

import fast

image_path = 'path to your WSI'

model = fast.DataHub().download('breast-tumour-segmentation-model')
importer = fast.WholeSlideImageImporter.create(image_path)

tissueSegmentation = fast.TissueSegmentation.create()\
.connect(importer)

generator = fast.PatchGenerator.create(
256, 256,
magnification=10,
overlapPercent=0
).connect(importer)\
.connect(1, tissueSegmentation)

segmentation = fast.NeuralNetwork.create(
model.paths[0] + '/pw_tumour_mobilenetv2_model.onnx',
scaleFactor=1./255.,
).connect(generator)

stitcher = fast.PatchStitcher.create()\
.connect(segmentation)

renderer = fast.ImagePyramidRenderer.create()\
.connect(importer)

heatmapRenderer = fast.HeatmapRenderer.create(channelColors={0: fast.Color.Green(), 1: fast.Color.Red()})\
.connect(stitcher)

fast.SimpleWindow2D.create()\
.connect(renderer)\
.connect(heatmapRenderer)\
.run()