Defining cutting-edge imaging informatics
Dr. Erickson's lab is engaged in developing innovative informatics tools that can extract and convey the wealth of information available in medical images in a clear and concise fashion to improve patient outcomes. This example automatically measures visceral and subcutaneous fat and also abdominal musculature, which is an important predictor of prognosis in several diseases as well as risk of vascular disease.
Overview
Mayo Clinic's Radiology Informatics Lab, directed by Dr. Bradley Erickson, is adding value to medical imaging by designing tools that:
- Assist in extracting quantitative and diagnostic information, including subtle textures that are not humanly perceptible
- Optimize the creation and quality of medical images, including deep-learning-based reconstruction
- Improve the workflow for planning, executing and reporting
The lab's main research focus is developing image-derived biomarkers of diseases. While our original work focused on brain cancer, the lab now works on a wide range of diseases and medical images, both within radiology and other image-producing departments.
In order to accomplish this, we have developed an image annotation tool that allows efficient segmentation of medical images. This tool has not only advanced image processing tools for segmentation, but can also execute external deep-learning tools that can produce segmentations that are then refined by the operator. The labels of structures are user defined, but include RadLex labels that enable interoperability.
The lab has also developed a data management tool that can accept DICOM images, remove PHI, and perform complex workflows that may include human and algorithmic steps. It also supports sophisticated versioning and has integrations with high-performance computing environments.
Affiliations
Dr. Erickson's Radiology Informatics Lab is affiliated with several Mayo Clinic research areas:
About Dr. Erickson
Bradley J. Erickson, M.D., Ph.D., is a professor of radiology at Mayo Clinic College of Medicine and Science in Rochester, Minnesota. A diagnostic radiologist, Dr. Erickson serves as the director of the Imaging Core within the Mayo Clinic Robert M. and Billie Kelley Pirnie Translational Polycystic Kidney Disease (PKD) Center. More recently, Dr. Erickson has been named medical director for Artificial Intelligence Innovation and Enablement across Mayo Clinic. Research in Dr. Erickson's lab focuses on optimally distilling vital information from medical images to improve patient outcomes.