Image Analysis: How To Surf the AI Wave
Weekend Course
ORGANIZERS: Akshay Chaudhari, Tolga Cukur, Yogesh Rathi, Gary Zhang
Saturday, 10 May 2025
311
13:00 -
17:00
Moderators: Mingming Wu & Mahmut Yurt
Skill Level: Basic to Advanced
Session Number: WE-13
No CME/CE Credit
Session Number: WE-13
Overview
This weekend session focuses on classical and transpiring AI techniques for image analysis tasks in MRI. The session will include spatial-temporal registration, tissue segmentation, disease classification and pathology/anomaly detection tasks. Conventional methods and new deep learning models will be comparatively surveyed to highlight their benefits in addressing these tasks.
Target Audience
Scientists, technologists, and clinicians interested in advanced image analysis techniques.
Educational Objectives
As a result of attending this course, participants should be able to:
• Describe the fundamentals and clinical applications of image analysis tasks such as registration, segmentation, and classification;
• List prominent examples of traditional and AI methods for performing image analysis tasks;
• Discuss practical issues related to algorithmic biases in AI models trained for image analysis; and
• Discuss practical issues related to performing AI-based image analysis across large-scale multi-site datasets with heterogeneous data distribution.
| 13:00 | | Clinical Role of Image Analysis in Radiological Workflows Yvonne Lui |
| 13:30 | | Primer on Machine Learning Techniques Mathews Jacob |
| 14:00 | | Registration: Fundamentals & Advanced AI Methods Wei Shao |
| 14:30 | | Segmentation: Fundamentals & Advanced AI Methods Julien Cohen-Adad |
| 15:00 | | Break & Meet the Teachers |
| 15:30 | | Estimation/Classification: Fundamentals & Advanced AI Methods Cem Deniz |
| 16:00 | | Emerging Machine Learning Methods for Data Analysis TBD |
| 16:30 | | Practical Considerations for Implementing AI-Based Analysis Amine Korchi Keywords: Image acquisition: Machine learning Practical Considerations for Implementing AI-Based Image Analysis in Radiology This session will explore the key technical and practical challenges of integrating AI-based image analysis into radiology workflows. Topics will include use-case/software selection, clinical validation and generalization, workflow integration, privacy and data security, interpretation challenges, performance monitoring, continual learning, and the integration of AI-generated results into final radiology reports. Emphasis will be placed on real-world implementation, addressing both the benefits and limitations of AI in clinical practice. This talk aims to provide a realistic, practice-driven perspective on the adoption and long-term sustainability of AI in radiology. |