Cancer Data Science Pulse

Imaging: A Key Component of a Cancer Data Ecosystem

Precision medicine has quickly moved to the forefront of clinical research and practice, and is particularly pertinent to cancer since cancer is a disease of the genome. The need to accelerate discovery in cancer research has been further propelled by the Beau Biden Cancer Moonshot, challenging the community to make a decade's worth of progress in five years. As part of the Moonshot, a Blue Ribbon Panel of experts convened to make recommendations on initiatives to accelerate cancer research, which included the creation of a National Cancer Data Ecosystem that "will enable all participants in cancer research and care communities to contribute, access, combine, and analyze diverse and inclusive data sets related to cancer." While a major focus in cancer research has initially been on genomics, it is clear that the diverse data types referenced by the Blue Ribbon Panel include a much broader set of data, including clinical, proteomic, and imaging.

Imaging can, and should, play a major role in cancer research, diagnosis, and treatment. As such, imaging data must be a key component of the Cancer Data Ecosystem described by the Blue Ribbon Panel.

The cancer research imaging community has long recognized the tremendous value it can bring to cancer research and clinical care. Radiological and pathologic imaging can play a complementary role to clinical and molecular data to offer key insights for diagnosis and treatment planning. But sharing imaging data and making it usable in a clinical setting has been a challenge. Traditionally, imaging data has been difficult to share and impossible to query. Images were stored in proprietary formats, and important data was recorded by clinicians in free text reports from which information could not be extracted.

For imaging to become integrated into the clinical care process and a major component of the Cancer Data Ecosystem, these challenges need to be overcome. Additional informatics capabilities must be developed.

Some efforts already underway to address these issues include the NCI Quantitative Imaging Network (QIN), whose goals are to promote the development of quantitative techniques for the measurement of tumor response to therapies in clinical care and clinical trial settings and to facilitate clinical decision making. Extracting measurable information in standard formats allows for reproducible and query-able results. Quantitative imaging will play a key role in this process of stratifying patients for personalized targeted treatments. Standards such as Digital Imaging and Communications in Medicine (DICOM), which allow for standardized transmission of images and associated data, and Annotation and Image Mark-up (AIM), which allows for knowledge to be added to an image in a clinical environment, have emerged to create more interoperability. Imaging has been identified as one the next critical data types to be incorporated into the Cancer Genomics Cloud Pilots.

Recognizing that a unified effort from the community will accelerate progress further, the NCI convened several workshops in the past three years, bringing together the leading experts in imaging informatics in cancer research. These workshops were designed to collaborate on the challenges and impediments to the integration of imaging into clinical practice, to identify the high priority informatics needs, and to highlight innovative work that was progressing the imaging community towards its goals.

A set of recommendations was developed from these workshops, along with examples of innovative projects, which have all been detailed in a White Paper entitled "Informatics Needs in Medical Imaging,"  authored by imaging informatics innovators from cancer centers across the country.

Some of the recommendations discussed include:

  • Using grant programs such as ITCR, QIN, and SBIR, as well as Grand Challenges, to encourage innovation.
  • Encouraging the adoption and further development of imaging data standards, and incentivize data sharing.
  • Working directly with clinicians, to educate them and to ensure tools will work in the clinical workflow.
  • Working with commercial partners, who have the ability to take a tool or method and bring it to market in a scalable, robust way.
  • Using open source tools and encouraging open source development, along with APIs that allow for interoperability.

We invite you to read the White Paper in its entirety and provide your thoughts and feedback. The White Paper is intended to be a living document, evolving as imaging informatics capabilities grow. We are moving towards a clinical workflow that includes imaging among the diverse data that provides critical input into treatment decisions. And we are progressing towards a Cancer Data Ecosystem that permits researchers, clinicians, and patients alike to view disease from a holistic, multi-domain perspective, allowing precision medicine to become a reality for each and every patient.

Graphic of Breast Imaging by Conventional PET displaying primary and axillary nodes.


For more information, or to provide your feedback, please contact Eve Shalley:

Edward Helton, Ph.D.
Government Sponsor, CBIIT Clinical Imaging Program
Robert Nordstrom, Ph.D.
Branch Chief, Imaging Guided Intervention Branch
Eve Shalley
program Manager, NCI CBIIT Cancer Informatics Branch
Older Post
Precision Medicine Inspires HPC
Newer Post
Democratizing Access to CCG Data: Cancer Genomics Cloud Pilots

Leave a Reply

Vote below about this page’s helpfulness.

Your email address will not be published.

Well said. Pretty nice article. Thank you for sharing.
Thank you for your comment. If you enjoyed this, please check out our Imaging Data Commons (IDC) page. IDC provides tools to search and visualize cancer imaging data, define cohorts, and use those cohorts for cloud-based analysis to better understand the disease and evaluate treatment options. Please let us know if there are other types of content you would enjoy seeing on this page!
Fascinating, the future is now!
Yes, it is! We’ll continue posting more information about cutting edge technologies, so stay tuned! If you enjoyed this, please check out our Imaging Data Commons (IDC) page. IDC provides tools to search and visualize cancer imaging data, define cohorts, and use those cohorts for cloud-based analysis to better understand the disease and evaluate treatment options.
Medical imaging is essential not only at initial diagnosis, but for monitoring how the disease is responding to treatment or if the disease is progressing, and when a treatment plan might be stopped or adjusted.
Medical imaging is absolutely critical in detecting and treating cancer. If you’re interested, the NCI Imaging Data Commons recently added 78 new, publicly available image collections. The platform also allows users to analyze 106 image and 6 analysis results collections directly from the IDC’s image viewer.
Interventional radiology is more complicated as it is image-guided surgery. They use the imaging tools mentioned above as a guide for their procedures.