This section of the documentation and code picks up after the slides have been added to the Specimen_Table_N.xlsx, a description of that file is located here, and and slides are transferred by the transferdeamon. After the transferdeamon module, all code assumes slides and file structure format in 4.6. have been adhered to, namely that the <ScanNN> subfolder is now in an additional im3 subfolder under each <SlideID>. Additional definitions on naming should be reviewed before continuing, found here.
In this section images are corrected for imaging warping and flatfield effects (5.7.). Next, cells are segmented and classified in each image using the multipass phenotype method in inForm. Each of these steps are run by custom code that process on perpetual loops with limited user interaction. An important step of this process is to create and check the quality control images generated on the cell classification output (that process is described in 5.8.6.) . After, images from all slides in a cohort pass the cell classification quality control, cell segmentation maps are rebuilt to remove cells not defined by the final merged cell segmentation (described here). Finally, slide macro regional annotations are applied to the images using the HALO software and distributed to the slides using a batch script 5.11.4.
A final checklist of things to verify and complete before loading into the database is added here. This section is a validatedata module which checks for any missing files, does the remaining file conversions, and provides final upkeep\ processing instructions.
- 5.1. Description
- 5.2. Contents
- 5.3. Instructions
- 5.4. Workflow Overview
- 5.5. Flatfield
- 5.6. Warping
- 5.7. Image Correction
- 5.8. VMinForm
- 5.9. Merge
- 5.10. Segmaps
- 5.11. Create & Transfer Annotations
- 5.12. Validate Data