How Will Big Data Improve Clinical and Basic Research in Radiation Therapy?

Abstract

Historically, basic scientists and clinical researchers have transduced reality into data so that they might explain or predict the world. Because data are fundamental to their craft, these investigators have been on the front lines of the Big Data deluge in recent years. Radiotherapy data are complex and longitudinal data sets are frequently collected to track both tumor and normal tissue response to therapy. As basic, translational and clinical investigators explore with increasingly greater depth the complexity of underlying disease processes and treatment outcomes, larger sample populations are required for research studies and greater quantities of data are being generated. In addition, well-curated research and trial data are being pooled in public data repositories to support large-scale analyses. Thus, the tremendous quantity of information produced in both basic and clinical research in radiation therapy can now be considered as having entered the realm of Big Data.

Figure 1

Most pressing issues in contemporary radiation oncology that could benefit from Big Data research. For each tumor site, areas in need of improvement in tumor and normal tissue outcomes are highlighted.

Figure 2

A cloud-based informatics infrastructure established for data transfer, quality assurance (QA) evaluation, data integration with NCI systems and standards. NCTN systems: OPEN - Oncology Patient Enrollment Network; RSS - Regulatory Support System; IROC – Image and Radiation Oncology Core. American College of Radiology (ACR) system: TRIAD – Transfer of Image and Data.

Figure 3

The Cancer Imaging Archive. (TCIA) is an NCI funded information repository that aggregates images (radiology, pathology), Radiation Therapy (RT) Information objects, annotations, clinical trial data, and information derived from quantitative image analysis to support Big Data analytics. TCIA has been extended to support curation, quality assurance, management and distribution of advanced RT information objects (illustrated in yellow) by incorporation of the POSDA (Perl Open Source Dicom) tool set (42) and associated processes.

Figure 4

Schematic of utilization of Oncospace at xxx. (a) Sources of raw data that are (b) “extracted, transformed and loaded” (ETL) into (c) Oncospace as metadata for efficient viewing and analysis by web-based query from (d) an outside institution as a member of the Oncospace consortium.

Figure 5

Knowledge Guided Radiotherapy. Big data will provide comprehensive knowledge for each patient from clinic to the laboratory. Such a knowledge-guided radiotherapy (KGRT) provides a potential to predict treatment outcome for each individual patient and guide personalized care for a maximized therapeutic gain for that individual and society (43). xxx is thanked for providing the initial formulation of this figure.