A.N. Pritzker Professor of Radiology, the Committee on Medical Physics, and the College
Vice-Chair for Basic Science Research, Dept. of Radiology
5841 South Maryland Avenue, MC2026
Chicago, Illinois 60637
Maryellen L. Giger is the A.N. Pritzker Professor of Radiology, the Committee on Medical Physics, and the College at the University of Chicago. She also serves as Vice-Chair for Basic Science Research in the Department of Radiology, University of Chicago. Involvement at the national and international level can be found in her CV. She served the University from 1998-2013 as Director of the CAMPEP-accredited Graduate Program in Medical Physics/ Chair of the Committee on Medical Physics.
Dr. Giger is considered one of the pioneers in the development of CAD (computer-aided diagnosis). For 30 years, she has conducted federally-funded research on computer-aided diagnosis, including computer vision and machine learning, in the areas of breast cancer, lung cancer, prostate cancer, lupus, and bone diseases. Her research in computational image-based analyses of cancer for risk assessment, diagnosis, prognosis, response to therapy has yielded various translated components, and she is now using these image-based phenotypes in imaging and multi-omics (e.g., genomics) association studies for cancer discovery and predictive modeling. She has authored or co-authored more than 400 scientific manuscripts (including 230 peer-reviewed journal articles), is inventor/co-inventor on approximately 30 patents, and serves as a reviewer for various national and international granting agencies, including the NIH and the U.S. Army.
Dr. Giger is a member of the National Academy of Engineering (NAE) of the National Academies. She has been awarded the William D. Coolidge Gold Medal from the American Association of Physicists in Medicine (AAPM), the EMBS Academic Career Achievement Award, the iBIOS iCON Innovator Award, and is a current Hagler Institute Fellow at Texas A&M University. She is a fellow of the American Institute for Medical and Biological Engineering (AIMBE), AAPM, SPIE, SBMR, and IEEE. In 2013, Professor Giger was named by the International Congress on Medical Physics (ICMP) as one of the 50 medical physicists with the most impact on the field in the last 50 years.
Dr. Giger serves as Editor-in-Chief of the SPIE Journal of Medical Imaging (JMI). In the recent past, she has served as the President and Treasurer of the AAPM and as President of SPIE. At the University, she is an active participant within the Cancer Center (UCCCC) and the Institute for Translational Medicine (ITM/CTSA).
Giger lab focuses on the development of multimodality CAD (computer-aided diagnosis), quantitative image analysis methods, and radiomics. Her research interests include digital medical imaging, computer-aided diagnosis, quantitative image analysis, and data-mining in breast imaging, chest/CT imaging, cardiac imaging, and bone radiography. The long-term goals of her research are to investigate, develop, and translate multi-modality quantitative image analysis techniques, which yield image-based tumor signatures and phenotypes, for improved cancer diagnosis, prognosis, patient care, and for discovering relationships between imaging and genomics, i.e., image-omics. Development of these methods includes novel means for lesion segmentation, and 2D, 3D, and 4D extraction of features characterizing the tumors and local background surround. These methods include development of computerized self-assessing lesion segmentation methods, which include methods for the computer to self-assess whether or not the lesion is well segmented as well as development of methods for incorporating extracted lesions features from multiple views and/or modalities, including those that weight features by the accuracy of the corresponding segmentation and those that use Bayesian neural networks with automatic relevance determination priors for joint feature selection and classification.
The Giger lab’s research also includes an investigation of the role of quantitative breast parenchymal characteristics in computerized analysis for both diagnosis and cancer risk assessment in an attempt to understand the relationship between image-based biomarkers and biological and clinical biomarkers. These methods include both supervised and non-supervised methods as well as deep learning approaches in machine learning.
Additional research involves methods for the optimization of the computer/human interface for presentation of computer output in CAD. Computer-determined estimates of the probability of malignancy of lesions are dependent on the prevalence of cancer in the training database, which most often does not correspond to the prevalence of cancer in the population from which the user has experience, e.g., the population seen in the user's medical practice. Thus, the user often has difficulty interpreting the computer-estimated probability of malignancy. Thus, the Giger lab is developing approaches with which to transform computer output to those that would match the internal parameters of the reader and thus provide useful indices of the probability of malignancy.
Dr. Giger also led the computer image-based, high-throughput MRI phenotyping analysis within NCI’s TCGA/TCIA breast image phenotype group. The group identified statistically significant associations between quantitative MRI radiomic features and various clinical, molecular, and genomic features in breast invasive carcinoma. Among the many novel findings were some highly specific imaging-genomic associations, which may be potentially useful in (a) imaging-based diagnoses that can inform the genetic progress of tumor and (b) discovery of genetic mechanisms that regulate the development of tumor phenotypes.
Overall, Dr. Giger’s research in image-based machine learning of breast cancer for risk assessment, diagnosis, prognosis, and response to therapy has yielded various translated components, and she is now using these image-based phenotypes, i.e., these “virtual biopsies” in imaging genomics association studies for discovery.
Dr. Giger is also a cofounder, equity holder, and scientific advisor for Quantitative Insights (QI), producers of QuantX, the first FDA-cleared, machine-learning driven system to aid in cancer diagnosis. QI was a finalist in the University of Chicago’s 2009-2010 New Venture Challenge.