Fakrul Islam Tushar

I am Fakrul Islam Tushar, PhD, an Assistant Professor at the University of Arizona, where I lead Tushar Lab. My research focuses on trustworthy AI for medical imaging, synthetic data, multimodal imaging, and virtual imaging trials, with an emphasis on building clinically grounded and reproducible AI systems for healthcare.

I received my PhD training at Duke University’s Department of Electrical & Computer Engineering and the Center for Virtual Imaging Trials (CVIT), where I worked on AI-driven medical imaging, simulation, and digital twin technologies. I also hold an Erasmus+ Master’s degree in Medical Imaging (MaIA), completed across Spain, Italy, and France, and a B.Sc. in Electrical Engineering from AIUB, where I graduated Cum Laude and received the Dean’s Award.

My work has been supported by Erasmus+ and Duke University and has been recognized through multiple honors, including the Best Poster Award at (Virtual Imaging Trials in Medicine 2024) and a Travel Award from the SPIE Medical Imaging Conference 2024. Beyond research, I have also been involved in professional and service organizations including IEEE, Teach For Bangladesh, and Literacy Through Leadership.

Open Source and Outreach Activities

I openly share most of the code developed during my research on GitHub. Explore my repository and additional resources below:

• HAID - Health AI Data Resource
• Nodule-Oriented Medical AI for Synthetic Imaging: NodMAISI
• Point-driven nodule segmentation: PiNS
• Virtual Lungs Screening Trials: VLST
• Context-Aware Nodule Augmentation: CaNA
• Duke Lung Nodule Dataset 2024: Zenodo
• AI in Lung Health Benchmark: GitHub
• In Silico Case-study: ReviCOVID