SVS preprocessor for Virtual Slide Viewer

Program that makes image pyramids from Aperio SVS files stored in Amazon S3. Also extracts metadata, label image, and thumbnail.

OpenSlide must be used to properly open SVS files.

General workflow for Virtual Slide Viewer deployments

1. Scanner dumps files onto ScanScope Workstation
2. aws s3 sync from ScanScope Workstation to images S3 bucket
   • delete local files on successful sync
3. Image preprocessor runs:
   • extract thumbnail and label images
   • decode slide ID from 2D barcode
   • extract metadata and put into DynamoDB slide table
   • generate Deep Zoom and IIIF views
   • upload views into viewer S3 bucket
4. Scanner technician reviews images for scanning errors
5. Pathologist reviews slides

Usage

$ alias aperio-proc='docker run --rm -ti -v ~/.aws:/root/.aws vanandelinstitute/aperio-proc'
$ aperio-proc -f "imageid.svs" -s "source-bucket" -d "dest-bucket" -t "dynamodb-table-name"