Segment features are numerical statistics that describe individual segments (e.g. cell size, or the mean or median pixel intensity for a marker or channel). You can select a CSV to import when setting up a new project or from the main menu. To import segment features from the main menu, navigate to
Segment Features and finally
For active image from CSV or
For project from single CSV . Note that these menu item will be disabled unless segmentation data has been loaded for the currently active image.
When importing segment features for a project Mantis expects the CSV file to have a header. The first column should contain the image folder name (the values in this column should match the folder names), the second column should contain segment/cell ids, and any remaining columns should contain cell/segment features. Mantis doesn’t use the column names/headers for the image folder name column or the segment/cell id column, but for the remaining columns Mantis uses the column name as the cell/segment feature name. For example, here are a few rows from a valid project CSV file:
Image Folder Name, Segment ID, Cell Size, CD4 Intensity set_one,220,1.23,3.45 set_one,707,2.34,4.32 set_two,77,4.45,5.43 set_two,140,6.54,7.68
When importing populations for a single image Mantis expects a similarly formatted CSV. The main difference is that first column (image folder name) is not included when importing for a single image. For example, here are a few rows from a valid image CSV file:
Segment ID, Cell Size, CD4 Intensity 220,1.23,3.45 707,2.34,4.32 77,4.45,5.43 140,6.54,7.68