Assistant Research Professor (Research Track), Radiology and Imaging Sciences

The Department of Radiology and Imaging Sciences at the University of Arizona in Tucson, AZ is seeking candidates for an open Assistant Research Professor (Research Track). The candidate will contribute to the department's artificial intelligence and data-driven medical imaging research program, with a primary focus on CT-based screening, diagnosis, and risk stratification. This research program integrates large-scale clinical imaging datasets, radiology reports, and virtual imaging trials to develop, validate, and translate AI tools that improve early disease detection, diagnostic precision, and patient outcomes. The successful candidate will assume an important role in NIH/NCI-funded research activities, including AI model development, simulation-based studies, and multi-disciplinary collaboration across radiology, engineering, and other clinical disciplines. The Assistant Professor may be required to travel to other sites to perform collaborative research and to attend conferences and meetings.

The University of Arizona has been recognized for our innovative work-life programs. For more information about working at the University of Arizona and relocation services, please click here.

• Participate in research on the development of artificial intelligence and data-driven methods for CT and other imaging modalities, including virtual imaging trials, synthetic data generation, and large-scale outcome modeling.
• Design, implement, and maintain end-to-end pipelines for curation, pre-processing, labeling, and management of large clinical imaging datasets and associated radiology reports.
• Participate in the development of protocols for in silico and clinical translational studies that leverage AI models to support screening, diagnosis, and patient risk assessment.
• Perform quantitative testing, validation, and quality control of AI models and simulation pipelines, including robustness, generalizability, and domain-adaptation studies across diverse datasets.
• Support clinical translation studies by partnering with technologists, study coordinators, data managers, and physicians to integrate AI tools and virtual imaging trials into clinical workflows.
• Program and test software algorithms and routines for image processing, image reconstruction and enhancement, deep learning model training and inference, explainability/visualization, and statistical analysis of AI performance.
• Conduct empirical evaluations of AI methods and virtual imaging trial results, benchmarking against clinical standards of care and existing imaging techniques.
• Interact with collaborators from within the institution, as well as collaborators from external and other organizations.
• Publish peer-reviewed scholarly works.
• Present at local, national and international conferences.
• Maintain an active research program.
• Teach as needed, advise and mentor students as needed.
• Serve on department and university committees.

Knowledge, Skills, and Abilities:

• Proficient and demonstrated skills, as evidenced by first-author publications, in designing and training deep learning models for CT and other medical imaging tasks, including weakly supervised learning and data-limited AI development.
• Proficient and demonstrated skills, as evidenced by first-author publications, in virtual imaging trials, synthetic data generation, and integration of simulated and clinical imaging data for screening and diagnostic applications.
• Proficient in scientific programming using Python and MATLAB, including experience with machine learning and data science libraries such as PyTorch, MONAI, TensorFlow, Scikit-learn, and related tools for 3D medical image analysis.
• Experience working with large-scale clinical imaging datasets and radiology reports, including data curation, NLP-based labeling, consensus labeling, and construction of benchmark datasets and open-access AI resources.
• Experience in GPU-accelerated computing and reproducible software development, including the use of containerization frameworks (e.g., Docker, Singularity) and collaborative code management (e.g., Git).
• Familiarity with methods for model interpretability, performance evaluation, and uncertainty assessment in AI for medical imaging, with an emphasis on transparent, trustworthy, and clinically relevant AI systems.

• Ph.D. in Electrical and Computer Engineering, Biomedical Engineering, Medical Physics, Computer Science, Applied Physics, or a closely related field, conferred by the start date.
• At least three years of research experience (which may include graduate and/or postdoctoral research) in one or more of the following areas:

1. Artificial intelligence / machine learning for medical imaging,
2. Computed tomography (CT) imaging
3. Virtual imaging trials or in silico studies
4. Synthetic data generation for imaging, or Large-scale clinical imaging data analysis.

• A record of peer-reviewed research productivity appropriate to the rank of Assistant Research Professor, including at least five first-authored publications in relevant areas such as AI for medical imaging, CT imaging, virtual imaging trials, or related quantitative imaging sciences.

• Five or more years of combined graduate and/or postdoctoral research experience in AI/deep learning for medical imaging, CT imaging, virtual imaging trials, or closely related fields.
• Demonstrated experience leading or coordinating multi-disciplinary research efforts, particularly projects that integrate radiology, engineering, and data science, and that involve large-scale clinical imaging datasets and radiology reports.
• Evidence of an active and externally visible research agenda, such as invited presentations, awards, competitive fellowships, leadership roles in collaborative projects, or open-access contributions (e.g., datasets, benchmarking studies, or software/toolboxes) in AI for CT, screening, or virtual imaging trials.
• Experience developing and evaluating end-to-end AI pipelines for medical imaging, including data curation, weakly supervised or label-efficient learning, model validation across diverse datasets, and methods to improve robustness, generalizability, and clinical relevance.