IEEE WIE ROC Virtual Speaker
Series
Trustworthy and Adaptive
AI for Imaging
Systems
Dr. Dimah Dera
Frederick and Anna B.
Weidman II Professor
Rochester Institute of
Technology
Abstract:
This work advances the
development of trustworthy and adaptive artificial intelligence for
imaging applications. Building on Bayesian uncertainty estimation and
probabilistic learning, the proposed framework enables imaging systems to
assess their confidence, adapt to changing environments, and maintain
reliability under data drift. Applications span medical image
segmentation, remote-sensing anomaly detection, and autonomous vision,
where uncertainty-guided adaptation enhances safety, interpretability,
and robustness. The approach integrates statistical signal processing
with deep learning to create imaging-based AI systems that are
self-aware, resilient, and transparent in real-world
conditions.
Speaker
Bio:
Dr. Dera holds a Ph.D.
and M.S. in Electrical and Computer Engineering and an M.A. in
Mathematics from Rowan University. Her research focuses on adaptive and
uncertainty-aware AI systems grounded in Bayesian theory and statistical
signal processing, with applications in healthcare, remote sensing, and
autonomous imaging. She has received numerous accolades, including the
NSF CISE CRII Award (2023), IEEE Best Paper Award (2019), and the NJ Tech
Council STEM Innovator to Watch Award (2019).
Intended
Audience:
Beginners, undergraduates, graduates. Those with
interest in the topic.<
/span>