Skip to content

image

Software for Medical Imaging

SMI contains several open-source applications and libraries for managing medical images, specifically DICOM. Our software has been designed, built, and validated in UK Trusted Research Environments (Safe Havens).

Note

This documentation repo is under early development. More information can be found in the documentation for each repository.

Site Contents

Our Software

Name Description
SmiServices Scale-able loading, linking and anonymisation of DICOM images for healthcare research environments (e.g. Safe Havens)
BadMedicine.Dicom CLI / Library for generating dicom files for use in testing applications. Images generated have 'realistic' tag data (based on aggregated tag data in dicom images taken in Scotland).
IsIdentifiable A tool for detecting identifiable information in data sources (CSV, DICOM, Relational Database and MongoDB)
DicomTypeTranslation FoDicom/FAnsiSql powered library for converting dicom types into database/C# types at speed.
RdmpDicom Plugin for RDMP that adds support for load, linking (with EHR data in relational databases) and extracting anonymous DICOM images for researchers.
dicompixelanon DICOM Pixel Anonymisation
DicomTemplateBuilder Windows application for building and testing DICOM templates
ctp-anon-minimal A minimal re-packaging of the RSNA MIRC Clinical Trials Processor (CTP), mainly providing the DICOMAnonymizer class
CogStack-SemEHR Semantic parsing of Electronic Health Records
DicomLoadScript
DqeToImagingJson
SCANDAN
StructuredReports DICOM Structured Reports
ansible Automated deployment of the SMI software stack
ctp-anon-cli A CLI tool to anonymise files using the RSNA MIRC Clinical Trials Processor (CTP) DICOM Anonymizer
db_analysis Various scripts used to query any monitor MongoDB
dicom-packager Scripts to package-up directories of DICOM files for transfer
dicom-standard The DICOM standard, in JSON format, parsed from the HTML version using Python scripts
dicom-to-minio Script to upload directories of DICOM files to MinIO
metacat Custom medical imaging metadata catalogue including classification based on DICOM metadata
nlp2phenome Infer patient phenotypes from identified named entities (instances of biomedical concepts)

Contact

Please get in touch at smi@epcc.ed.ac.uk for all enquiries.