AI Selection Without the AI Sales Pitch: How a Leading Radiology Center Benchmarked Multiple Radiology AI Tools with One Evaluation Framework
.png)
About
When it comes to radiology AI, performance on paper rarely tells the full story. In this webinar, a head radiologist shares how their team tested multiple AI solutions from different vendors on the same data, with the same clinicians, using one structured, vendor-neutral evaluation framework.
Learn how they moved beyond demos and marketing claims to generate their own clinical evidence—capturing accuracy, usability, and workflow fit in real-world conditions. Whether you're evaluating tools for procurement, piloting new use cases, or validating in-house models, this is your inside look at a scalable, clinician-led approach to AI selection.
Agenda
Why radiology AI selection needs real-world validation
Explore why accuracy alone isn’t enough and how performance can vary across sites, patient populations, and workflows.
Inside the evaluation: one use case, multiple vendors, same data
See how a leading hospital structured a head-to-head comparison across radiologists and solutions, without local installs or vendor coordination.
What vendor-neutral evaluation enables that demos can’t
Understand how a single, platform-based approach removes bias, scales across departments, and supports both regulatory and strategic goals.
Speakers
Priv. Doc. Dr. Peter Brader is a board-certified radiologist specializing in molecular and oncologic imaging with consideration of the use of AI. He studied medicine at the University of Graz (MD, 2001) and completed his radiology training at the Medical University of Graz and Memorial Sloan Kettering Cancer Center in New York (2003–2009). In 2010, he received his habilitation in radiology with a thesis on the use of Escherichia coli Nissle 1917 as a vector to enhance tumor detection via PET and optical imaging. His scientific and clinical focus areas include urogenital radiology, imaging biomarkers, and AI. Since 2013, Dr. Brader has been part of Diagnostikum Group. In 2020, he was appointed Medical Director of Diagnostikum Linz, where he combines leadership responsibilities with active clinical practice.
Julia Moosbauer is a mathematician with a PhD in data science from the Munich Center for Machine Learning (MCML), where she focused on enhancing the transparency and understandability of Automated Machine Learning (AutoML) systems. In 2019, she co-founded deepc, a leading company in medical artificial intelligence (AI). As the COO, she led deepc to achieve ISO 13485 and ISO 27001 certifications, ensuring compliance with the highest standards in information security, data protection, and quality management. Today, deepc’s platform, deepcOS®, empowers radiology facilities from across the world to seamlessly integrate AI models into their diagnostic workflows, advancing the healthcare system responsibly.
