FAIR-compliant clinical, radiomics and DICOM metadata of RIDER, interobserver, Lung1 and head-Neck1 TCIA collections

摘要: Purpose: One of the most frequently cited radiomics investigations showed that features automatically extracted from routine clinical images could be used in prognostic modeling. These have been made publicly accessible via The Cancer Imaging Archive (TCIA). There have been numerous requests for additional explanatory metadata on following datasets — RIDER, Interobserver, Lung1, and Head–Neck1. To support repeatability, reproducibility, generalizability, transparency in research, we publish subjects’ data, features, digital imaging communications medicine (DICOM) headers these four with descriptive metadata, order to more compliant findable, accessible, interoperable, reusable (FAIR) data management principles. Acquisition validation methods: Overall survival time intervals were updated using a national citizens registry after internal ethics board approval. Spatial offsets primary gross tumor volume (GTV) regions interest (ROIs) associated Lung1 CT series improved the TCIA. GTV open-source Ontology-Guided Radiomics Analysis Workflow (O-RAW). We reshaped output O-RAW map extraction settings latest version Radiomics Ontology, so as consistent Image Biomarker Standardization Initiative (IBSI). Digital imaging was extracted research Semantic DICOM (SOHARD, GmbH, Fuerth; Germany). Subjects’ data described Radiation Oncology Ontology. All the above published Resource Descriptor Format (RDF), is, triples. Example SPARQL queries are shared reader use online triples archive, which intended illustrate how exploit this submission. Data format: accumulated RDF through SPARQL endpoint where archived. endpoint is remotely queried graph database web application at http://sparql.cancerdata.org. queries intrinsically federated, such can efficiently cross-reference clinical, DICOM, within single query, while being agnostic original format coding system. federated work same way even if partitioned across multiple servers dispersed physical locations. Potential applications: public availability resources replication, reproducibility studies by academic community. The example may freely modified readers depending their research question. interoperability reusability supported referencing existing ontologies. readily findable aforementioned link. Scripts used to create available code repository linked submission: https://gitlab.c om/UM-CDS/FAIR-compliant_clinical_radiomics_and_DICOM_metadata