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

Andre Dekker,Rianne Fijten,Matthijs Sloep,Martijn P.A Starmans,Ananya Choudhury,Peter Feltens,Detlef Grittner,Leonard Wee,Hugo Aerts,Johan Van Soest,Petros Kalendralis,Zhenwei Shi,Rene Monshouwer,Stefan Klein,Ivan Zhovannik,Alberto Traverso

doi:10.1002/mp.14322

Abstract

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

Highlights

Clinical radiological imaging, such as computed tomography (CT), is a mainstay modality for diagnosis, screening, intervention planning, and follow-up for cancer patients worldwide.[1]
Aerts et al showed that CT radiomics features by themselves could contain information that is potentially prognostic of overall survival in nonsmall cell lung (NSCLC) and head-and-neck (HN) cancer.[3]
The core collection being made publicly available here consists of (a) improvements to the four clinical imaging datasets described in the seminal radiomics publication by Aerts et al.,[3] (b) extracted radiomics features described in line with image biomarker standardization initiative (IBSI) recommendations,[27,31] and (c) updates to the subject clinical data associated with the aforementioned image collections

Summary

INTRODUCTION

Clinical radiological imaging, such as computed tomography (CT), is a mainstay modality for diagnosis, screening, intervention planning, and follow-up for cancer patients worldwide.[1]. Regardless of complexity, can be written out in full as a series of machine-readable sentences consisting of strictly three pieces; subject (start vertex) — predicate (edge) — object (end vertex) Such “triples” are the basis of the resource descriptor format (RDF) that is a type of universal data storage format on the World Wide Web. Machine-based data mining and inferencing tasks are feasible in a highly efficient manner, being simplified to a “pattern matching” problem. Machine-based data mining and inferencing tasks are feasible in a highly efficient manner, being simplified to a “pattern matching” problem The objective of this open data submission is to stimulate studies into repeatability, reproducibility, replication, and reusability of radiomics features from multiple datasets. The core collection being made publicly available here consists of (a) improvements to the four clinical imaging datasets described in the seminal radiomics publication by Aerts et al.,[3] (b) extracted radiomics features described in line with IBSI recommendations,[27,31] and (c) updates to the subject clinical data associated with the aforementioned image collections

ACQUISITION AND VALIDATION METHODS

DISCUSSION

CONCLUSIONS