Advanced Certificate curriculum
The RSNA Imaging AI Certificate Program offers a pathway of certificate courses, providing you with the ability to harness the AI knowledge critical to meeting the challenges in the medical imaging field.
The curriculum described on this page is for the program’s Advanced Certificate course. It’s the next step on the pathway to help radiologists confidently use AI tools and is designed to provide a deeper understanding of the steps involved in using AI algorithms in medical imaging.
Outcomes and learning objectives
Upon completion of the six-module curriculum, enrollees will earn the Advanced Certificate, recognizing their ability to evaluate the fairness of AI models across populations, interpret the AI lifecycle, examine the pitfalls of using AI in a clinical setting and recognize the impact of the regulatory environment affecting the use of AI in medical imaging.
Advanced Certificate course learning objectives:
- Prepare radiologists, physicists, data scientists and clinical researchers to evaluate the AI model’s fairness across various populations.
- Interpret the AI lifecycle beginning with training and test data curation to FDA approval.
- Provide participants with a deep understanding of the pitfalls of dataset curation, pre-processing and annotation when initiating AI for clinical use.
- Recognize the impact of regulatory environment, the clinical AI marketplace and ethical considerations on the delivery of AI in health care.
Each case-based module allows you to learn at your own pace through a series of pre-recorded videos and a variety of hands-on activities that build on concepts established in the previous modules.
Module 1: Introduction to AI
You will assess the purpose and complexities involved when developing machine learning models and applications. This module will demonstrate different approaches to generating models and resolving AI issues. You will also learn to identify the ethical considerations involved in data sharing including patient privacy and consent.
Module 2: Dataset Curation, Image Preprocessing, and Annotation
You will identify key elements of image annotation tools and the infrastructure required and interpret the standard formats most suited to imaging annotations. This module will help you recognize pertinent factors in DICOM metadata, pixel level de-identification and existing software solutions. You will also assess the benefits and limitations of using human readers to annotate medical images and computer-based analyses.
Module 3: Model Building
Module 4: Model Evaluation
Module 5: FDA Clearance and Marketplace Considerations
Module 6: AI Ethics
“ I teach informatics and will use this knowledge to enhance my course content. ”
— Advanced Certificate Enrollee, 2023
This activity has not been designated for continuing medical education credit.