On October 24th from 15:00 - 16:00, Dr. Binod Bhattarai at University of Aberdeen (UK) will hold a talk on the following subject:
Towards Trustworhy AI in Medical Imaging
Artificial Intelligence is rapidly changing how we see and understand the human body through medical imaging. From detecting hidden tumors to generating clinical reports, AI holds incredible promise for transforming healthcare. But for these technologies to truly make an impact, they must earn our trust.
In this talk, Dr. Binod Bhattarai will explore the three key pillars of trustworthy AI: privacy, reliability, and explainability. He will share his team’s latest research in federated learning, which allows hospitals to collaboratively train powerful AI models without sharing sensitive patient data, and out-of-distribution (OOD) detection, which helps systems recognize when they encounter rare or uncertain cases and explain their decisions. Dr. Bhattarai will also introduce a new benchmark for detecting hallucinations in vision-language models used in gastrointestinal imaging—a critical step toward making multimodal medical AI more factual and dependable.
Through these innovations, the talk highlights how we can move toward AI systems that are not only intelligent but also transparent, safe, and worthy of trust in real-world healthcare.
The webinar will be hosted by Arezo Shakeri, member of the Student Council and PhD candidate at the University of Stavanger.
Register to join