Icon

GOEBEL-Berwick-CellPose-Seg-Measurement-Categorisation

Mixed cells - segmentation and classification
The workflow processes all the image files located inthe specified in the first node. The workflow assumes that: - Channel 1 is IgG (green)- Channel 2 is Cy5 (red)- Channel 3 is the channel to be segmented(transmitted light) - Channel 4 is DAPI (blue)- further channels can be present but will be ignoredIf the channels are not in the specified order listedabove, the workflow will still happen, i.e. it will not fail,but the results will not be correct!If the channels are not arranged as listed above, the Optional: Generate images from segmentation Good idea for quality control purposes generate file names for output files Measure feature (object) properties, e.g. size (numbers of pixels), intensity-related parameters (Mean, Median, Maximum, ...) This list can be expanded. Prepare data for export and write feature (object) properties to disk. The first column contains the filename for therespective objects found. The Label column contains a unique identifier (number 1-n) for each object per image. Specify the file name and destination for the results file wich can then be processed further. Best practise for saving a workflow: Reset all the nodes. Otherwise it will takeup much more space on the disc and willbe slower to save/load. Any specified folders for input and outputwill remain even after resetting all thenodes. Display results as violin plots after the results table has been tidied externally. Node 1: specify folder containing images for processingNode 2: read all image filesNode 3Node 4: segment cells in TL channelNode 6: extract transmitted light channel for segmentationNode 7: save segmentation image (optional but highly recommended for quality control)Node 9Node 21: export resultsas csv fileNode 22: add column with filename to results tableNode 26: generate new Row ID columnNode 27: re-generate original rowID for each imageNode 28: generatge segmentation file nameNode 29Node 30Node 31Node 32: rename columns corresponding to the channel signalNode 33Node 39: split multichannel imageNode 40Node 74: loop through fluorescence channelsNode 76: join segmentation and fluorescence channelNode 81: aggregate results in one tableNode 82: measure intensity parameters for each identified object for each channel and fileNode 83: filter for fluorescence channelsNode 84: remove duplicate columnsNode 85: calculate object areas in numbers of pixelsNode 86Node 87Node 89: cell type classifierNode 91: plot paramters for classified cell typesNode 92Node 93: label objects with object numberNode 95Node 97: plot parametersper conditionNode 98: plot parametersfor conditions, determinedfrom file namesNode 99Node 101: read tidied results.csv fileNode 107Node 109: read in classification sizeand intensity cutoffparametersNode 118 List Files/Folders Image Reader(Table) Path to String CellposeSegmentation (CPU) Column Filter Image Writer Column Filter CSV Writer Joiner RowID String Manipulation String Manipulation String Manipulation Column Resorter Joiner Column Rename Labeling to Image Splitter(uncropped) Split CollectionColumn Column ListLoop Start Joiner Loop End (ColumnAppend) FeatureCalculator (BETA) Column Filter Column Filter Image SegmentFeatures Joiner Column Rename Rule Engine Color Manager Violin Plot(Plotly) String Manipulation Row Filter String Manipulation Color Manager Violin Plot(Plotly) CSV Reader Calculate rawintegrated intensities CSV Reader Classification rulevariable setup The workflow processes all the image files located inthe specified in the first node. The workflow assumes that: - Channel 1 is IgG (green)- Channel 2 is Cy5 (red)- Channel 3 is the channel to be segmented(transmitted light) - Channel 4 is DAPI (blue)- further channels can be present but will be ignoredIf the channels are not in the specified order listedabove, the workflow will still happen, i.e. it will not fail,but the results will not be correct!If the channels are not arranged as listed above, the Optional: Generate images from segmentation Good idea for quality control purposes generate file names for output files Measure feature (object) properties, e.g. size (numbers of pixels), intensity-related parameters (Mean, Median, Maximum, ...) This list can be expanded. Prepare data for export and write feature (object) properties to disk. The first column contains the filename for therespective objects found. The Label column contains a unique identifier (number 1-n) for each object per image. Specify the file name and destination for the results file wich can then be processed further. Best practise for saving a workflow: Reset all the nodes. Otherwise it will takeup much more space on the disc and willbe slower to save/load. Any specified folders for input and outputwill remain even after resetting all thenodes. Display results as violin plots after the results table has been tidied externally. Node 1: specify folder containing images for processingNode 2: read all image filesNode 3Node 4: segment cells in TL channelNode 6: extract transmitted light channel for segmentationNode 7: save segmentation image (optional but highly recommended for quality control)Node 9Node 21: export resultsas csv fileNode 22: add column with filename to results tableNode 26: generate new Row ID columnNode 27: re-generate original rowID for each imageNode 28: generatge segmentation file nameNode 29Node 30Node 31Node 32: rename columns corresponding to the channel signalNode 33Node 39: split multichannel imageNode 40Node 74: loop through fluorescence channelsNode 76: join segmentation and fluorescence channelNode 81: aggregate results in one tableNode 82: measure intensity parameters for each identified object for each channel and fileNode 83: filter for fluorescence channelsNode 84: remove duplicate columnsNode 85: calculate object areas in numbers of pixelsNode 86Node 87Node 89: cell type classifierNode 91: plot paramters for classified cell typesNode 92Node 93: label objects with object numberNode 95Node 97: plot parametersper conditionNode 98: plot parametersfor conditions, determinedfrom file namesNode 99Node 101: read tidied results.csv fileNode 107Node 109: read in classification sizeand intensity cutoffparametersNode 118List Files/Folders Image Reader(Table) Path to String CellposeSegmentation (CPU) Column Filter Image Writer Column Filter CSV Writer Joiner RowID String Manipulation String Manipulation String Manipulation Column Resorter Joiner Column Rename Labeling to Image Splitter(uncropped) Split CollectionColumn Column ListLoop Start Joiner Loop End (ColumnAppend) FeatureCalculator (BETA) Column Filter Column Filter Image SegmentFeatures Joiner Column Rename Rule Engine Color Manager Violin Plot(Plotly) String Manipulation Row Filter String Manipulation Color Manager Violin Plot(Plotly) CSV Reader Calculate rawintegrated intensities CSV Reader Classification rulevariable setup

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: mheldb
Created at: 2021-03-15
On NodePit since: 2021-09-18
Last update: 2024-04-18
Created with KNIME version: v4.3.3
Tags: fluorescencesegmentationCellpose

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!