The AI Theatre: Keynote

AI-SC 1 - Generative AI in clinical radiology

Lectures

1
Chairperson's introduction

Chairperson's introduction

03:00Saif Afat, Tübingen / DE, Lisa C. Adams, Munich / DE

2
Generative AI in clinical radiology: current status, challenges, and future directions

Generative AI in clinical radiology: current status, challenges, and future directions

20:00Marc Kohli, San Francisco / US

3
Energy, water, and hidden costs: sustainability impacts of (generative) AI in radiology

Energy, water, and hidden costs: sustainability impacts of (generative) AI in radiology

20:00Florence Xini Doo, Baltimore / US

4
Q&A: Should AI Model Cards carry energy labels and/or kg CO2 equivalent metrics?

Q&A: Should AI Model Cards carry energy labels and/or kg CO2 equivalent metrics?

17:00Q&A: Should AI Model Cards carry energy labels and/or kg CO2 equivalent metrics?

Waste-free Wednesday
3 min
Chairperson's introduction
Saif Afat, Tübingen / Germany
Lisa C. Adams, Munich / Germany
20 min
Generative AI in clinical radiology: current status, challenges, and future directions
Marc Kohli, San Francisco / United States
  1. To learn about the current applications of generative AI in clinical radiology, interpretive and non-interpretive.
  2. To appreciate the challenges associated with integrating generative AI into clinical radiology practice, such as regulatory aspects, monitoring strategies and open-source vs third party considerations.
  3. To understand the future directions of generative AI in radiology, focusing on upcoming innovations, potential clinical benefits, and strategies to overcome existing barriers for broader adoption.
20 min
Energy, water, and hidden costs: sustainability impacts of (generative) AI in radiology
Florence Xini Doo, Baltimore / United States
  1. To learn methods for estimating energy and water consumption associated with the use of AI technologies in radiology departments.
  2. To appreciate the importance of understanding resource usage to promote sustainability in the integration of AI within radiology practices.
  3. To understand the necessity for ongoing research into the energy consumption of AI in radiology, and how this compares to the anticipated clinical and operational benefits.
17 min
Q&A: Should AI Model Cards carry energy labels and/or kg CO2 equivalent metrics?

This session offers AI-generated subtitles.