Keyvan Farahani, Ph.D.

Biomedical Imaging Informatics Specialist

  • Current Role

    Current Role

    I support NCI’s cancer data science with emphasis on imaging informatics. My expertise lies in knowledge of biomedical imaging, minimally invasive cancer therapies, the development of data repositories, and software tools for machine learning/artificial intelligence applications in imaging and cancer research. I am also on a part-time detail with the National Heart, Lung, and Blood Institute (NHLBI), where I support the development of imaging and artificial intelligence platforms for key NHLBI data science initiatives, including the BioData Catalyst. I am an active member of the American Association of Physicists in Medicine (AAPM) and the Society for Imaging Informatics in Medicine (SIIM).

  • Projects

    Projects

    Stylized image of a brain as viewed from the top. Instead of brain structures the brain is portrayed as a series of interconnected lines and dots.

    Artificial Intelligence in Medical Imaging

    Focuses on both imaging data AI-readiness and the curation of harmonized tools/models in a central repository.
    Stylized photo of a scientist peering at a computer screen showing different views of a brain image. The doctor is wearing a lab coat, surgical mask, surgical gloves, and surgical cap. To the left of the image is another graphic showing a locked padlock against a background of interconnected lines and dots.

    Medical Image De-identification Initiative (MIDI)

    Addresses and implements best practices for the protection of patient privacy in clinical image sharing.
    Photo shows a female scientist in a laboratory setting. She’s looking at brain images on a computer screen.

    Medical Imaging Challenge Cloud Infrastructure (MedICCI)

    Benchmarks imaging analysis tools using annotated data sets.
    Stylized image of a cloud showing repositories of data represented by circles with small icons hovering over top. The repositories include canine, imaging, genomic, and other types of data.

    NCI Cancer Research Data Commons (CRDC)

    Connects researchers with data sets and tools to analyze large- scale cancer research data in the cloud.
  • Publications

    Publications

    View Keyvan’s PubMed Summary

    Featured Publications

    Methods for Segmentation and Classification of Digital Microscopy Tissue Images. Frontiers in Bioengineering and Biotechnology, 2019.

    Advancing The Cancer Genome Atlas Glioma MRI Collections with Expert Segmentation Labels and Radiomic Features. Scientific Data, 2017.

    The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS). IEEE Transactions on Medical Imaging, 2014.

  • In the News

    In the News

    NCI’s Imaging Data Commons Releases New Type of Image Data
    NCI’s Imaging Data Commons has released several notable updates, including the release of three collections by the Human Tumor Atlas Network, which comprise a new multichannel fluorescence imaging data set.
    Share Feedback on NCI’s Imaging Data Commons
    Do you work with imaging data and tools? Share feedback on NCI’s Imaging Data Commons!
    Deadline Extended: Submit Abstracts for the Eighth Computational Approaches for Cancer Workshop by August 16
    As the necessity for and availability of large data sets in cancer applications grows, so do the challenges when conducting research and clinical applications with computational solutions. Share your experience with how to address such challenges.
  • Background

    Background

    Education

    • B.S., Physics, Sonoma State University in California
    • Ph.D., Biomedical Physics, University of California at Los Angeles

    Previous Work

    • Program Director for Image-Guided Interventions, NCI Cancer Imaging Program

    Recent Awards

    • Fellow, The American Institute for Medical and Biological Engineering (AIMBE)
    • Fellow, The American Association of Physicists in Medicine (AAPM)