Radiology and Imaging Sciences
Research
Integrating intelligent imaging, computation and translational science to accelerate clinical breakthroughs
Research scientists in the Department of Radiology and Imaging Sciences are redefining what’s possible in medicine by uniting intelligent imaging, advanced AI and bold scientific imagination.
Our teams drive a continuous cycle of discovery — transforming fundamental imaging science into next-generation diagnostics and therapeutics designed to elevate patient care. This translational mindset fuels a future where imaging not only reveals disease but anticipates it, guides treatment and shapes precision health.
Through deep collaboration with basic science and clinical partners across the University of Arizona, the University of Arizona Cancer Center and the College of Medicine – Tucson, our investigators help lead imaging-driven clinical trials that push the boundaries of innovation. These partnerships empower us to design and evaluate AI-enabled imaging studies that accelerate the journey from breakthrough insight to real-world clinical impact.
Our Researchers
Stephen Adamo, PhD
The Attention, Detection and Medical Observation (ADAMO) Lab, led by Stephen Adamo, PhD, investigates human visual search in complex medical environments to improve diagnostic accuracy, utilizing in silico (simulated) images to analyze how radiologists and trainees interpret data. The team specifically examines the impact of imaging modalities (2D vs. 3D) and artificial intelligence systems on observer performance, aiming to translate these insights into clinical practice.
- Visual search in medicine: Studies how observers, from students to professionals, navigate complex medical images to identify anomalies.
- Imaging modalities and AI: Analyzes the influence of 2D vs. 3D imaging and AI systems on diagnostic decisions.
- Expertise analysis: Compares search patterns across different experience levels, including residents and radiologists.
- Fundamental principles: Uses in silico simulated images to translate cognitive research into clinical applications.
Research Highlights
This project uses electroencephalography (EEG) to identify when and why cancerous lesions are missed in mammographic images, revealing neural markers of detection failures to improve diagnostic accuracy.
Using behavioral measures and eye tracking, this work quantifies how breast density impacts lesion detection, identifying mechanisms that contribute to missed cancers in dense tissue.
This project examines how observers of varying expertise search for cancer in 2D versus 3D imaging, revealing how tomosynthesis changes visual search patterns and detection performance.
This work evaluates how clinicians interact with black-box and explainable AI tools, revealing how different forms of AI guidance influence visual search, decision-making and diagnostic performance.
This project investigates how AI design, including anthropomorphic features, shapes user trust and diagnostic behavior, informing how AI systems can be integrated safely and effectively into clinical workflows.
FDA Center for Devices and Radiological Health (CDRH)
Maria Altbach, PhD
Research led by Maria Altbach, PhD, advances novel MRI techniques aimed at improving image quality, quantitative accuracy and clinical disease detection.
- Radial MRI acquisition: Developing highly undersampled radial imaging strategies that deliver motion‑robust images with high spatial and temporal resolution.
- Parametric mapping: Creating high‑resolution quantitative mapping methods to more precisely characterize disease.
- Disease detection: Collaborating with industry and clinical partners to enhance early detection and treatment of conditions such as liver and colorectal cancers.
Lars Furenlid, PhD
Research led by Lars Furenlid, PhD, advances molecular and medical imaging through integrated innovation in hardware, software and contrast‑agent technologies.
- Molecular imaging hardware: Designing next‑generation detectors and instrumentation for SPECT, PET and CT.
- Human brain and preclinical systems: Developing specialized SPECT platforms to support drug discovery in cancer and neurodegenerative diseases.
- Adaptive imaging: Creating multi-modal systems where data from one modality (MRI, CT) dynamically guides another for precise functional imaging.
- Image science: Building GPU-accelerated tools for data inversion, tomographic reconstruction and image quality evaluation.
Hsin Wu (Andy) Tseng, PhD
The research of Hsin Wu (Andy) Tseng, PhD, focuses on medical image reconstruction for computed tomography (CT), radiation dosimetry including Monte Carlo simulations, and task-based system optimization to advance early cancer detection and image-guided interventions.
- Advanced reconstruction: Developing deep-learning and compressed-sensing algorithms to reduce image noise and reconstruct high-quality from sparse-view data or incomplete datasets.
- Dual-energy imaging: Developing dual-energy cone-beam CT methods to deliver superior tissue differentiation for oncological imaging and real-time, image-guided surgical procedures.
- Dose optimization: Employing mathematical model observers to safely minimize patient X-ray exposure while maintaining maximum diagnostic image quality.
- Breast CT systems: Collaborating on a novel, simultaneous bilateral breast CT imaging system designed for ultra-fast, high-resolution imaging of both breasts for breast cancer detection and diagnosis.
Fakrul Tushar, PhD
The Tushar Lab, led by Fakrul Islam Tushar, PhD, advances data‑centric artificial intelligence for healthcare, integrating clinical, simulated and synthetic datasets to build trustworthy, rigorously validated medical AI systems.
- Data‑centric healthcare AI: Leveraging weak supervision and human-AI collaboration to curate large‑scale medical datasets and improve model reliability.
- Virtual imaging trials: Using digital twins and in silico patient cohorts to enable reproducible, simulation‑based evaluation of clinical workflows.
- Generative modeling: Producing synthetic medical images to overcome small‑sample limitations and train more robust diagnostic models.
- Thoracic diagnostic tools: Developing AI systems for lung cancer screening, thoracic imaging and automated lung‑nodule detection.
Srinivasan Vedantham, PhD, DABR, FAAPM
The research group led by Srinivasan Vedantham, PhD, the BIG‑CT Laboratory, advances the design, development and clinical translation of next‑generation X‑ray imaging systems and tomographic technologies.
- Breast CT innovations: Developing compression-free, high‑resolution 3D breast CT technology for ultra-fast scans aimed at improving comfort, eliminating tissue overlap, and enabling more accurate breast cancer screening and diagnosis.
- AI-based image reconstruction: Developing advanced image reconstruction methods for cone-beam CT using deep learning methods.
- Breast density assessment: Creating methods for precise breast‑density quantification to improve breast cancer risk estimation.
- Advanced detector technology: Advancing interventional imaging through improved reconstruction algorithms and next‑generation detector designs.
Russell Witte, PhD
Led by Russell Witte, PhD, the Experimental Ultrasound and Neural Imaging (EUNIL) Lab explores ultrasound imaging — with a twist.
- Ultrasound imaging integrating light, sound and electricity
- Biomedical applications from head to toe
- Epilepsy, stroke, Parkinson’s, cancer, arrythmias, tendinopathy
Research Highlights
University of Arizona / ARPA-H Project (2025-present)
Principal Investigator
Co-Investigators
Business Lead: Ken Coffey
Junior Investigator: Mitchell Bartlett, PhD
Patient Ambassador: John Black (BRIDGE)
University of Arizona / NIH / Flinn Foundation Project (2014-Present)
Technical Team
University of Arizona
- Nan-kuei Chen, PhD, Department of Biomedical Engineering
- Leonid Kunyansky, PhD, Department of Mathematics
University of Washington
- Matthew O’Donnell, PhD, Department of Bioengineering
University of Michigan
- Zhen Xu, PhD, Department of Biomedical Engineering
Brain Science Team
University of Arizona Department of Psychology
Clinical Team
Visiting Scholar
- Real-time US + PA imaging at depths of ~10 mm and resolution of ~0.1 mm
- Intrinsic contrast: blood vessels, blood oxygen, melanin, lipid, water, collagen
- Skin cancer: classify cutaneous lesions/melanomas, monitor treatment
Clara N. Curiel-Lewandrowski, MD
Other team members
- Eric Reichel, PhD
- Chris Salinas
- Abhiman Gupta
- Flatfoot deformity prevalent in women more than 50 years of age
- Early diagnosis allows for early and less invasive therapy
- Solution: 3D imaging of structure and elastic properties of foot ligaments
Team members
Weimin Zhou, PhD
The Computational Imaging and Visual Intelligence Laboratory (CIVIL) Lab, led by Weimin Zhou, PhD, aims to advance both human and machine visual perception while improving diagnostic accuracy and interpretive efficiency in clinical imaging.
- Medical image synthesis: Developing deep generative models to create realistic medical images for training, augmentation and algorithm evaluation.
- Image quality assessment: Using AI/ML-driven observer models to perform objective, task-based assessments that maximize the clinical value of imaging systems.
- Signal detection and visual search: Studying how clinicians and algorithms detect abnormalities within complex images to streamline diagnostic workflows.
- Deep learning reconstruction: Designing reconstruction networks optimized for clinical task performance, image clarity and efficient interpretation.