When: PM, Sunday, September 16, 2018
The objective of this tutorial is to introduce the MICCAI community to the new kinds of DICOM objects and capabilities that can be used for storage and communication of the data typically produced in the process of quantitative image analysis (such as image segmentation results, measurements, radiomics features, parametric maps).
Over the course of presentations and hands-on sessions, we explain how DICOM can, and we believe should, be used for storing your processing results such as segmentations, parametric maps and volumetric measurements.
After completing this tutorial, attendees are expected to develop an understanding of the relevant new and long-established capabilities of the DICOM standard, as well as some of the tools that they can use to experiment with adoption of the standard in their everyday research.
This talk will introduce the main concepts of the DICOM standard, the role it plays in the modern healthcare enterprise, its overall purpose and capabilities.
This talk will discuss the relevance of the DICOM standard in quantitative imaging research, motivate its use and discuss specific capabilities as applied to the storage of common image analysis outputs, such as image segmentation results, parametric maps and region-based quantitative measurements.
An overview, capabilities and summary of experience with the various data management platforms that can work with the DICOM data for medical image computing applications. Some of the tools we plan to cover will include XNAT and dcm4chee.
Attendees will be guided over the steps of installing the free open source software tools that will be used in the tutorial, and which can help the attendees work with the DICOM data in their everyday work after the conference. The tools utilized in the tutorial will include 3D Slicer, MITK, DCMTK, Atom, and dcmqi. We will also distribute the sample dataset that will be used in the tutorial.Closer to the event we will provide instructions on how to install the prerequisites for your platform in advance of the tutorial!
In this hands-on exercise attendees will use one or more open source tools (3D Slicer, MITK, dcmqi) to load DICOM data, segment regions of interest, and extract radiomics features. The results will be stored using standard (no private attributes) DICOM objects. The resulting standard-compliant data storage will be contrasted with the alternative approaches used in the community.
In this session we will go over the steps of extracting various data attributes (both in the source images, and in the research results stored in DICOM representation) from the standard DICOM representations of the analysis results, generating summary visualizations and preparing representations (e.g., CSV spreadsheets) suitable for the analysis by non-DICOM-enabled tools.
Radiologist, medical informaticist, DICOM open source software author and editor of the DICOM standard. Formerly the co-chair of the IHE Radiology Technical Committee and industry co-chairman of the DICOM Standards Committee.
Assistant Professor in Radiology at the Brigham and Women's Hospital Department of Radiology, and Harvard Medical School. Together with Ron Kikinis, Andrey is a co-PI of the Quantitative Image Informatics for Cancer Research project.
Medical Imaging Interaction Toolkit (MITK) team lead at the German Cancer Research Center. His work focuses on research enabling technologies, interoperability, open source and application-oriented research in radiology, surgery and radiation oncology.
DICOM4MICCAI tutorial first took place during MICCAI 2017 in beautiful Quebec City, Canada.
If interested, we welcome you to check out the archived event page from DICOM4MICCAI 2017 that includes downloadable slides, pointers to the related tools, and hands-on materials presented at that event.