Mark Anastasio, PhD

Academic Department Radiology
Research Program Oncologic Imaging
Mark Anastasio

Researcher Background

Primary Academic Title

R. Gilbert Jost Distinguished Professor of Imaging Sciences and Artificial Intelligence

Training

  • 1995 - 2001: Research Assistant, Department of radiology, University of Chicago, Chicago, IL

Education

  • 1992: BS, Illinois Institute of Technology, Chicago, IL
  • 1993: MSE, University of Pennsylvania, Philadelphia, PA
  • 1995: MS, University of Illinois at Chicago, Chicago, IL
  • 2001: PhD, University of Chicago, Chicago, IL

Research & Select Publications

Research Interest

My research focuses on advancing computational and AI-enabled imaging science and establishing rigorous methods for designing, reconstructing, and evaluating biomedical images. A central theme is understanding how imaging systems and AI methods preserve or alter information relevant to scientific and clinical tasks. My group develops image reconstruction and inverse-problem methods, task-based image quality assessment techniques, learning-based observer models, and approaches to characterize potentially misleading features in

AI-generated images. Applications in oncologic imaging include photoacoustic and ultrasound computed tomography, virtual imaging studies of breast imaging systems, computational and quantitative imaging, and methods for evaluating the clinical utility and reliability of AI-based image reconstruction and analysis. Our overarching goal is to help translate innovative imaging and AI technologies into reliable tools for cancer research and patient care.

Research Publications

A virtual imaging framework for three-dimensional quantitative optoacoustic tomography using stochastic numerical breast phantoms.
Authors:

Park S, Jeong G, Villa U, Anastasio MA

Journal & Year:

Photoacoustics • 2026

Application of a virtual imaging framework for investigating a deep learning-based reconstruction method for 3D quantitative photoacoustic computed tomography.
Authors:

Cam RM, Park S, Villa U, Anastasio MA

Journal & Year:

Photoacoustics • 2026

Approximating the ideal observer for joint signal detection and estimation tasks by the use of Markov-Chain Monte Carlo with generative adversarial networks.
Authors:

Li D, Li K, Zhou W, Anastasio MA

Journal & Year:

J Med Imaging (Bellingham) • 2025

Investigating the use of signal detection information in supervised learning-based image denoising with consideration of task-shift.
Authors:

Li K, Li H, Anastasio MA

Journal & Year:

J Med Imaging (Bellingham) • 2024