Warning! OUTDATED BROWSER DETECTED!   Please update your browser immediately for a better experience on this website. Learn More
  • To:
  • Quantitative Imaging Accelerates Clinical Research

    May 01, 2014

    New technology and sophisticated imaging software are hastening the progress of quantitative imaging as a clinical research tool.

    Technological breakthroughs in medical imaging hardware and increasingly sophisticated image processing software are hastening the progress of quantitative imaging as a clinical research tool—a particularly critical application for cancer therapy, experts say.

    “Across all modalities—from nuclear medicine to ultrasound, CT and MR imaging, medical imaging provides a rich amount of quantitative data,” said Katarzyna J. Macura, M.D., Ph.D., associate professor of radiology, urology and oncology at The Johns Hopkins University in Baltimore. “Radiologists perform some type of quantitative imaging in their everyday practice, from measuring abnormalities to assessing blood flow, tissue perfusion or diffusion and tracer uptake. This quantitative approach to imaging data allows more objectivity and facilitates comparisons between different states of the same disease or different disease processes.”

    As defined by RSNA’s Quantitative Imaging Biomarkers Alliance (QIBA), quantitative imaging is “the extraction of quantifiable features from medical images for the assessment of normal or the severity, degree of change, or status of a disease, injury, or chronic condition relative to normal.”

    Quantitative imaging provides a noninvasive method for characterizing therapy response, said Sonia Pujol, Ph.D., a researcher at Brigham and Women’s Hospital, Harvard Medical School in Boston and director of training of the National Alliance for Medical Imaging Computing (NAMIC) and the Neuroimaging Analysis Center. Dr. Pujol, along with Dr. Macura and Ron Kikinis, M.D., a professor of radiology at Brigham and Women’s Hospital, Harvard Medical School and principal investigator of the 3D slicer, presented a workshop on quantitative imaging at RSNA 2013. Dr. Pujol said quantitative imaging data can potentially be used to give radiologists complementary information for interpreting images.

    “Biomarkers—as measurable characteristics used to indicate a biological state and where imaging can contribute quantifiable data—are used in oncologic clinical trials, for example, as surrogate end-points to evaluate treatment response, and especially to detect early failure of potentially toxic treatments or to predict patient outcome,” said Macura who received a 2003 Toshiba America Medical Systems/RSNA Research Resident Grant.

    “Most radiologists are familiar with the World Health Organization (WHO) and Response Evaluation Criteria in Solid Tumors (RECIST) criteria that were introduced to enable meaningful comparison from one exam to another and from one clinical trial to another,” Dr. Pujol said. “Continuous advances are being made in imaging technology and post-processing software tools, and the intersection of quantitative and clinical sciences offers great promise to help improve outcome prediction and tumor response to therapy.”

    Quantitative imaging has the potential to provide useful clinical information in situations where disease progress is difficult to assess, Dr. Pujol said. “For example, small changes in slowly evolving pathologies such as meningiomas, can be clinically significant but difficult to detect in longitudinal MR imaging scans. Quantitative imaging methods that combine input from medical experts with advanced image analysis tools can aid clinicians in treatment decisions.”

    Radiologists May Be Hesitant to Adopt Quantitative Imaging

    Still, quantitative imaging has a ways to go to reach its full potential.

    For example, one study shows that radiologists may be somewhat resistant to using quantitative imaging. A study by Richard Abramson, M.D., and colleagues published in the July 2012 issue of Magnetic Resonance Imaging analyzed MR imaging and CT reports from two randomly selected weekdays at a mixed academic-community practice for the presence of quantitative descriptors. They found that while 44 percent of all reports contained at least one quantitative metric, just 2 percent contained an advanced quantitative metric.

    Quantitative imaging is an active field of research, and its translation into routine clinical radiology practice has been hindered by many factors, including a lack of standardization, Dr. Pujol said.

    “There’s a need to validate quantitative imaging biomarkers by evaluating the variability among methods and comparing different methods with outcomes,” Dr. Pujol said. “Similar problems exist with diffusion tensor imaging (DTI) tractography, and we are trying to evaluate current tractography techniques and define standards to ascertain quality features for surgical guidance through the DTI Challenge.”

    There are also workflow issues, she explained. “Quantitative imaging methods can be time consuming and require the integration of additional post-processing tools to the clinical workflow,” Dr. Pujol said.

    Consortia such as the National Cancer Institute’s Quantitative Imaging Network (QIN) and QIBA are trying to improve the value of quantitative imaging by working on the challenges associated with standardizing and validating quantitative imaging biomarkers and post-processing techniques, Dr. Pujol said. “Validation is the key to accelerating the transfer of innovative imaging methods developed in research to the clinics.”

    RSNA Workshop Gives Radiologists Practical Experience

    The RSNA workshop taught by Drs. Pujol, Macura and Kikinis—“Quantitative Medical Imaging for Clinical Research and Practice”—is designed to provide an introduction to quantitative imaging through a series of case studies involving different organs and multiple imaging modalities, Dr. Pujol said.

    The course combines the presentation of quantitative imaging biomarkers for diagnosis as well as clinical trial outcome measures with hands-on sessions on the basics of quantitative measurements.

    “The hands-on aspect of the course is intended to give radiologists practical experience on the latest methods developed in medical research, as well as an opportunity to give feedback on quantitative imaging tools still in early development,” Dr. Pujol said.

    “This combination of practical experience and the opportunity to influence how quantitative imaging tools are deployed makes this workshop an invaluable experience for radiologists interested in improving the value of imaging in clinical practice,” said RSNA Science Advisor Daniel Sullivan, M.D., a professor in the Department of Radiology at Duke University and chair of the QIBA Steering Committee.

    The course also provides presenters with the opportunity to give radiologists practical experience in quantitative image analysis using 3D Slicer—a multi-platform open-source software package for medical image computing and 3D visualization supported by a multi-institution effort and several large-scale grants funded by the National Institutes of Health. As a clinical research tool, 3D Slicer brings clinicians and research scientists together in the prototyping, development and evaluation of novel image analysis tools, Dr. Pujol said.

    “For example, 3D Slicer is currently used by several NCI Quantitative Imaging Network sites in a variety of clinical research applications, such as pharmacokinetic analysis of prostate DCE MR imaging and PET/CT analysis for therapy response assessment in head and neck cancer,” she said.

    Such hands-on courses and tutorials—including the RSNA quantitative imaging course—can hopefully accelerate the transfer of novel post-processing tools to clinical researchers, Dr. Pujol said.

    Web Extras

    Katarzyna J. Macura, M.D., Ph.D.
    Daniel Sullivan, M.D.
    Sonia Pujol, Ph.D.
    Post-processing of a recurrent/residual anaplastic olygoastrocytoma case within the 3D Slicer software
    Post-processing of a recurrent/residual anaplastic olygoastrocytoma case within the 3D Slicer software. Right, top: Co-registered T2-weighted image, (middle) T1-weighted image and (bottom) Mean Diffusity map with segmented edema (light blue) and tumor regions (yellow, pink). Left: 3D surface models of tumor and edema reconstructed from the segmented regions. Image courtsey of Sonia Pujol, Ph.D., and Alexandra Golby M.D.
    comments powered by Disqus