Fakrul Islam Tushar, PhD

Assistant Professor

Dr. Tushar is an Assistant Research Professor in the Department of Radiology and Imaging Sciences at the University of Arizona. He earned his PhD in Electrical and Computer Engineering from Duke University in 2025, where he conducted research in health AI and generative AI at the Center for Virtual Imaging Trials (CVIT). Prior to his doctoral studies, he served as a Research Associate at Duke University Medical Center from 2019 to 2021. He holds an MSc in Medical Imaging and Applications through the Erasmus Mundus Joint Master (MaIA) program and a BSc in Electrical and Electronic Engineering from the American International University-Bangladesh (AIUB).

Dr. Tushar directs the Tushar Lab, a newly established research group advancing data-centric AI for healthcare and medicine through data discovery, intelligent tools, and the integration of clinical, simulated, and synthetic datasets. His work spans large-scale data curation using weak supervision and human-AI collaboration, virtual imaging trials for reproducible in-silico evaluation, and generative modeling for synthetic medical image creation. The lab's overarching goal is to develop trustworthy, rigorously evaluated, and clinically grounded AI systems for healthcare and medical imaging.

Dr. Tushar's research has been published in leading venues including Medical Image Analysis, Radiology: Artificial Intelligence, SPIE Journal of Medical Imaging, and Artificial Intelligence in Medicine, with presentations at RSNA and SPIE Medical Imaging. He is committed to open and reproducible science; his toolkits, datasets, and benchmarking resources are publicly available through his GitHub and lab website. He is the recipient of the Erasmus Mundus Joint Master Scholarship and has been recognized with Best Poster awards at the International Summit of Virtual Imaging Trials in Medicine (2024) and the Duke All Pratt Poster Competition (2022).

Research Interests: 
  • Data-Centric AI & Weak Supervision
  • Trustworthy AI Evaluation for Medical Imaging
  • Foundation Model Benchmarking
  • Virtual Imaging Trials & In Silico Evaluation
  • Generative AI & Medical Image Synthesis
Degrees
  • PhD: Electrical and Computer Engineering, Duke University, 2025
  • MSc: Medical Imaging and Applications (Erasmus Mundus), University of Girona, Spain, 2019
  • BSc: Electrical and Electronic Engineering, American International University-Bangladesh, Dhaka, 2017
Honors and Awards
  • Best Poster Presentation, International Summit of Virtual Imaging Trials in Medicine, 2024
  • Best Poster, All Pratt Poster Competition, Duke University, 2022
  • Erasmus Mundus Joint Master Scholarship, European Union, 2017-2019
  • Master’s Thesis Grant, Duke University Medical Center, 2019
  • Academic Honour “Cum Laude,” American International University-Bangladesh, 2017
  • Dean’s Award for Undergraduate Final Project, 2016
  • Merit Scholarship, American International University-Bangladesh, 2013-2017

Select Publications

2026

Tushar, F. Islam, L. Dahal, S. Sotoudeh-Paima, E. Abadi, W. P. Segars, J. Y. Lo, and E. Samei, "Utility of the virtual imaging trials methodology for objective characterization of AI systems and training data.", J Med Imaging (Bellingham), vol. 13, issue 1, pp. 014506, 2026 Jan. PMCID: PMC12906637 

2025

Wang, A. J., F. Islam Tushar, M. R. Harowicz, B. C. Tong, K. J. Lafata, T. D. Tailor, and J. Y. Lo, "The Duke Lung Cancer Screening (DLCS) Dataset: A Reference Dataset of Annotated Low-Dose Screening Thoracic CT.", Radiol Artif Intell, vol. 7, issue 4, pp. e240248, 2025 Jul. PMCID: PMC12319698  PMID: 40237601
Tushar, F. Islam, L. Vancoillie, C. McCabe, A. Kavuri, L. Dahal, B. Harrawood, M. Fryling, M. Zarei, S. Sotoudeh-Paima, F. Chi Ho, et al., "Virtual lung screening trial (VLST): An in silico study inspired by the national lung screening trial for lung cancer detection.", Med Image Anal, vol. 103, pp. 103576, 2025 Jul. PMCID: PMC12147717  PMID: 40209556

2022

Tushar, F. Islam, V. M. D'Anniballe, R. Hou, M. A. Mazurowski, W. Fu, E. Samei, G. D. Rubin, and J. Y. Lo, "Classification of Multiple Diseases on Body CT Scans Using Weakly Supervised Deep Learning.", Radiol Artif Intell, vol. 4, issue 1, pp. e210026, 2022 Jan. PMCID: PMC8823458  PMID: 35146433
D'Anniballe, V. M., F. Islam Tushar, K. Faryna, S. Han, M. A. Mazurowski, G. D. Rubin, and J. Y. Lo, "Multi-label annotation of text reports from computed tomography of the chest, abdomen, and pelvis using deep learning.", BMC Med Inform Decis Mak, vol. 22, issue 1, pp. 102, 2022 Apr 15. PMCID: PMC9011942  PMID: 35428335