Daniel Marcus, Ph.D.

Associate Professor

Biomedical Informatics and Data Science Program
Neurosciences Program

  • dmarcus@wustl.edu

Research Abstract:

1. Development of computationally-derived imaging biomarkers
We use research and clinical imaging data, primarily MRI, PET, and CT, to develop biomarkers of diseases that impact the structure and function of the brain. Areas of interest include Alzheimer’s disease, stroke, and brain tumors. We use various analytic tools, including deep learning and other artificial intelligence techniques, to assess brain function and disease state. In our research on metastatic brain tumors, for example, we are developing tools to automatically segment lesions, calculate imaging-based features of these lesions, and predict how they will respond to image-guided radiation therapy. In our research on stroke, we are developing predictive models of how the stroke will progress and identify which patients will require surgical intervention.

2. Clinical translation of quantitative neuroimaging applications
The imaging biomarkers developed in the above research theme will often have direct clinical application. We are developing tools to bring these applications into clinical workflows. This work includes comprehensive clinical validation of each application as well as creation of the clinical reports and imaging workstation software used by clinicians to incorporate the applications into their clinical routines. In one such project, we have collaborated with the Department of Neurosurgery to deploy a brain imaging application that maps the location of functional brain
regions in patients with glioblastomas. The surgeons use these maps to guide surgical resection of these tumors. The application has been used in over 2000 surgical patients to date and is now being prepared for FDA approval.

3. Neuroinformatics software and services
To enable the above work, we develop software and infrastructure to support big data and massive processes. The open source XNAT platform is developed in the lab and used by institutions around the world to enable neuroimaging research and clinical translation. The lab also leads the informatics program for the Human Connectome Project (HCP) and the National Cancer Institute’s Integrative Imaging Informatics for Cancer Research (I3CR) program. We also oversee a number of data sharing initiatives, including HCP and OASIS Brains. Together, the data from these initiative have been used in over 1500 publications.

Selected Publications:

Marcus DS, Olsen TR, Ramaratnam M, Buckner RL. The Extensible Neuroimaging Archive Toolkit: an informatics platform for managing, exploring, and sharing neuroimaging data. Neuroinformatics. 2007 Spring;5(1):11-34. PubMed PMID: 17426351.

Schwartz Y, Barbot A, Thyreau B, Frouin V, Varoquaux G, Siram A, Marcus DS, Poline JB. PyXNAT: XNAT in Python. Front Neuroinform. 2012;6:12. PubMed PMID: 22654752; PubMed Central PMCID: PMC3354345.

Archie KA, Marcus DS. DicomBrowser: software for viewing and modifying DICOM metadata. J Digit Imaging. 2012 Oct;25(5):635-45. PubMed PMID: 22349992; PubMed Central PMCID: PMC3447088.

Herrick R, McKay M, Olsen T, Horton W, Florida M, Moore CJ, Marcus DS. Data dictionary services in XNAT and the Human Connectome Project. Front Neuroinform. 2014;8:65. PubMed PMID: 25071542; PubMed Central PMCID: PMC4080781.
Back To Top

Follow us: