SCC Outputs
Page Contents
SCC has two main output types: image files csv files.
The image files SCC outputs will be annotated copies of the images used as inputs. These are generated by Batch Mode (with Make Output Images selected) and Co-expression Analysis (with Add Contours to Output Images selected). Preview Mode also displays a set of images that reflect how the input image was processed.
csv (comma-separated values) files can be viewed and modified in any spreadsheet editor. SCC uses these to compile data for many images during Batch Mode and Co-expression Analysis.
Note: csv is a very basic file format. Visual and formula spreadsheet editor features can be used while the file is open, but only the text values in each cell will be saved upon closing.
Preview Mode analyzes one image and outputs a detailed visual overview of how SCC processed the image. Use it to tune parameters to a dataset by checking a few images for accurate cell identification.
Interpretation of each image:
- Source
- The input image in grayscale. Brightness will be inverted if Fluorescent is selected.
- Truncated
- Pixels brighter than the mode background pixel value have been made equivalent to that value, reducing noise.
- After Threshold
- An adaptive binary threshold has been applied to the image.
- For every pixel, the average brightness of the surrounding pixels has been measured, and pixels darker than that average by a certain amount (determined by Selection Strength) were made black. All other pixels are white.
- After Gap Filling
- Regions of black pixels have been expanded, then contracted.
- This serves to resolve some of the ring-like nature of punctate protein expression, though SCC can still have problems with dim punctate expression.
- After Size Filtering
- Regions of black pixels have been contracted, then expanded.
- This erases any regions which are smaller than the minimum radius of cells (determined by Size) and and erases isthmuses (thin connectors between otherwise large regions).
- This image is directly used to make the contours in the Final image.
- Final
- Contours have been drawn and filtered, and the cell counts displayed on the top left.
- Cyan contours are counted cells, Yellow are contours that failed the circularity filter (determined by Circularity Threshold), Pink are contours that failed the area filter (determined by Minimum and Maximum Area).
These images are the same as the Final output image from Preview Mode
Batch mode reports a summary of the cell counts and data from each image it processed.
Interpretation of each column:
- Image
- Full path of image file.
- Count
- Number of counted cells in the image.
- Background
- The brightness of the background tissue.
- Equivalent to the mode pixel value of the image.
- Spread
- The difference in brightness of the background and an estimation of the cells in the image.
- Can be used to tune Selection Strength.
- Average Area
- Average area of counted cells in the image.
- Can be used to tune Minimum and Maximum Area.
The calculated mean and standard error of the mean are entered at the bottom of each column.
If Batch Mode is run with Spread Only selected, the Counts and Average Areas columns will be filled with "N/A".
These images are the same as the Final output image from Preview Mode, with co-expressing cells additionally outlined in white.
Note: Co-expression is reported for each image, meaning images from separate channels can be cross-checked to ensure overlapping cells were identified from the perspective of both channels.
Co-expression Analysis reports the overlapping contours in each image across all selected channels.
Interpretation of each column:
- Image
- Full path of image file.
- Count
- Number of counted cells that appeared on all selected channels.
- Average Area
- Average area of counted cells that appeared on all selected channels.
Note: Co-expression is reported for each image, meaning images from separate channels can be cross-checked to ensure overlapping cells were identified from the perspective of both channels. (Notice rows 9 and 30 in the image are of the same tissue and have different counts)