ARRS Clinical Artificial Intelligence in Radiology 2026

Videos + eBook | Clinical AI • Deep Learning • LLMs • Generative AI • Multimodal AI • AI Governance

Develop a practical understanding of how artificial intelligence is transforming radiology with ARRS Clinical Artificial Intelligence in Radiology 2026, a comprehensive educational program from the American Roentgen Ray Society (ARRS).

Designed to move radiologists from basic AI awareness toward practical clinical implementation, this course covers the foundations of artificial intelligence as well as advanced topics including:

• Deep Learning
• Natural Language Processing – NLP
• Large Language Models – LLMs
• Generative AI
• Agentic AI
• Radiomics
• Multimodal Foundation Models
• AI Workflow Automation
• AI-Assisted Image Interpretation
• AI Governance
• Regulation
• Ethics
• Bias and Fairness
• Human-AI Collaboration
• AI Research and Education

The program also explores real-world AI applications across major radiology subspecialties, including breast imaging, neuroradiology, abdominal imaging, musculoskeletal imaging, pediatric radiology, cardiothoracic imaging, interventional radiology, and nuclear medicine.

A central theme throughout the course is how radiologists can integrate AI into clinical practice while maintaining human-centered, radiologist-led healthcare.

Course Details

• Course: Clinical Artificial Intelligence in Radiology
• Provider: American Roentgen Ray Society – ARRS
• Year: 2026
• Original Course Dates: April 12–13, 2026
• Original Event: ARRS 2026 Annual Meeting
• Location: Pittsburgh, Pennsylvania
• Specialty: Radiology / Artificial Intelligence
• Modules: 7
• Format: Video-Based Educational Course
• Included Content: Videos + eBook
• eBook: 182 Pages / 24 Chapters
• Language: English
• Primary Focus: Clinical Artificial Intelligence in Radiology

Course Directors

Shandong Wu, PhD

University of Pittsburgh

Yee Seng Ng, MD

University of Washington

Hyun Soo Ko, MD

Peter MacCallum Cancer Centre

Course Overview

Artificial intelligence has moved beyond experimental radiology research and is increasingly becoming part of real-world clinical imaging.

AI tools can now contribute to:

• Image interpretation
• Workflow optimization
• Worklist prioritization
• Detection and segmentation
• Quantitative imaging
• Clinical decision support
• Report analysis
• Natural language processing
• Research
• Education
• Prognostic prediction
• Multimodal data integration

However, successful AI adoption requires more than simply purchasing an algorithm.

Radiologists need to understand:

How AI Works → What Problem It Solves → How It Is Validated → How It Is Deployed → How Performance Is Monitored → How It Affects Patients and Radiologists

ARRS Clinical Artificial Intelligence in Radiology 2026 provides a structured framework for navigating these questions.

Learning Outcomes

After completing this course, learners should be better able to:

• Analyze the fundamentals of artificial intelligence and its value in radiology.
• Understand major AI concepts relevant to medical imaging.
• Explain deep learning and modern AI architectures.
• Understand applications of natural language processing in radiology.
• Describe the role of large language models in imaging practice.
• Understand generative AI and emerging agentic AI systems.
• Review applications of radiomics.
• Create a practical framework for AI implementation in clinical practice.
• Understand important considerations when deploying AI algorithms.
• Evaluate AI governance and regulatory requirements.
• Recognize ethical and legal considerations surrounding clinical AI.
• Understand multimodal foundation models.
• Review AI applications across major radiology subspecialties.
• Understand approaches to AI model development and research.
• Explore how AI can support radiology education.
• Recognize algorithmic bias and fairness issues.
• Understand effective radiologist-AI collaboration.
• Maintain a human-centered approach to AI-assisted healthcare.

Module 1: Getting to Know AI

Overview of Radiology Artificial Intelligence: Latest Progress

Speaker: Tessa Cook, MD, PhD

An overview of the rapidly evolving field of artificial intelligence in radiology and recent developments affecting clinical imaging.

Topics include:

• Evolution of radiology AI
• Current AI capabilities
• Clinical applications
• AI adoption
• Emerging technologies
• Future directions

Primer on Artificial Intelligence

Deep Learning, NLP, Large Language Models, Generative AI, Agentic AI & Radiomics

Speaker: Hyun Soo Ko, MD

This foundational lecture provides an introduction to important terminology and technologies behind modern medical AI.

Deep Learning

Review the principles behind neural networks and deep-learning systems used in medical imaging.

Applications may include:

• Classification
• Detection
• Segmentation
• Image reconstruction
• Quantitative imaging
• Prognostic modeling

Natural Language Processing – NLP

NLP enables computers to analyze and generate human language.

Potential radiology applications include:

• Radiology report analysis
• Information extraction
• Report classification
• Clinical communication
• Structured reporting
• Research database creation

Large Language Models – LLMs

Large language models have introduced new possibilities for:

• Report generation
• Summarization
• Clinical information retrieval
• Education
• Decision support
• Communication
• Workflow automation

Generative AI

Generative models can create new:

• Text
• Images
• Synthetic data
• Structured information

The course explores their potential applications and limitations in radiology.

Agentic AI

Agentic artificial intelligence represents an emerging generation of systems capable of performing multi-step tasks and interacting with tools, software, and clinical workflows.

Radiomics

Radiomics extracts large quantities of quantitative information from medical imaging.

Potential applications include:

• Tumor characterization
• Prognostic prediction
• Treatment-response assessment
• Phenotyping
• Personalized medicine

AI Can Improve Radiology Workflow Efficiency by Automating Noninterpretive Tasks

Speaker: Linda Moy, MD

Not every AI application needs to interpret images.

Artificial intelligence can also automate time-consuming workflow tasks surrounding image interpretation.

Possible areas include:

• Worklist management
• Scheduling
• Protocol support
• Prior examination retrieval
• Report workflow
• Communication
• Follow-up management
• Administrative tasks

Reducing repetitive tasks may allow radiologists to spend more time on clinical interpretation and patient care.

Artificial Intelligence to Improve Radiology Imaging Interpretation

Speaker: Shandong Wu, PhD

Reviews how AI can directly support diagnostic imaging interpretation.

Applications can include:

• Abnormality detection
• Lesion identification
• Segmentation
• Quantification
• Classification
• Computer-aided diagnosis
• Risk prediction
• Decision support

The objective is not simply to automate interpretation but to understand how AI can augment radiologist performance.

Module 2: AI Clinical Implementation

Understanding AI algorithms is only one component of clinical adoption.

Implementation requires consideration of:

• Clinical workflow
• Governance
• Validation
• Regulation
• Legal responsibility
• Patient safety
• Monitoring
• Algorithm maintenance

Legal and Ethical Considerations in AI Implementation

Speaker: Julian Rivera, JD

Artificial intelligence introduces new legal and ethical questions into medical practice.

Important areas include:

• Clinical responsibility
• Liability
• Patient privacy
• Data ownership
• Algorithm transparency
• Informed decision-making
• Accountability
• Safety
• Ethical implementation

Artificial Intelligence Deployment

Speaker: Tessa Cook, MD, PhD

This lecture moves from theoretical AI performance to practical clinical implementation.

A potential deployment pathway can be considered as:

Clinical Need → Product Evaluation → Local Validation → Workflow Integration → Deployment → Performance Monitoring → Maintenance

Important questions include:

• Does the algorithm solve a meaningful clinical problem?
• Is performance validated in the relevant patient population?
• How will it integrate with PACS and existing systems?
• What happens when the algorithm fails?
• Who monitors performance?
• How are software updates managed?

AI Regulation and Governance

A Practice Perspective on Assessment, Deployment and Maintenance of AI Algorithms

Speaker: Melissa Davis, MD, MBA

Effective AI governance is essential for safe clinical deployment.

The lecture addresses issues including:

• Algorithm assessment
• Governance committees
• Clinical validation
• Deployment approval
• Performance monitoring
• Algorithm updates
• Maintenance
• Risk management
• Regulatory compliance

Module 3: Going Beyond Images to Multimodality

Modern AI increasingly integrates more than one type of medical information.

Future clinical systems may combine:

• Imaging
• Clinical notes
• Laboratory results
• Pathology
• Genomics
• Demographics
• Previous examinations
• Structured health-record data

This multimodal approach may allow AI systems to develop richer clinical context.

Medical Imaging Dataset Curation for Artificial Intelligence

Speaker: Heather Whitney, PhD

High-quality AI begins with high-quality data.

Topics include:

• Dataset construction
• Annotation
• Ground truth
• Data quality
• Labeling
• Dataset diversity
• Bias
• Training data
• Validation data
• Generalizability

Poorly constructed datasets can produce poorly generalizable algorithms regardless of model sophistication.

Multimodal Foundation Models in Radiology

Speaker: Christian Bluethgen, MD

Foundation models represent an important evolution in medical artificial intelligence.

Rather than building separate models for every individual imaging task, foundation models may learn broad representations that can support multiple downstream applications.

Multimodal models may integrate:

Images + Text + Clinical Data + Other Medical Information

Potential applications include:

• Clinical reasoning support
• Image-text integration
• Report generation
• Patient-specific analysis
• Research
• Decision support

Physics and Artificial Intelligence in CT

Speaker: Lifeng Yu, PhD

Explores interactions between AI and computed tomography physics.

Relevant applications may include:

• CT reconstruction
• Image quality
• Noise reduction
• Dose optimization
• Acquisition
• Quantitative imaging
• Automated processing

Module 4: AI Use Cases in Radiology Subspecialties

Breast Imaging Artificial Intelligence

Speaker: Constance Lehman, MD, PhD

Reviews AI applications in breast imaging, potentially including:

• Mammography
• Breast cancer detection
• Risk prediction
• Workflow support
• Screening
• Diagnostic assessment

Artificial Intelligence in Neuroradiology

Speaker: Paulo Kuriki, MD

Explores AI applications in:

• Brain imaging
• Stroke
• Neuroimaging segmentation
• Detection
• Quantitative analysis
• Clinical decision support

Artificial Intelligence in Abdominal Imaging

Speaker: Yee Ng, MD

Reviews AI applications involving abdominal radiology.

Potential areas include:

• Liver imaging
• Pancreatic imaging
• Abdominal oncology
• Organ segmentation
• Quantitative biomarkers
• Opportunistic imaging
• Disease detection

Current and Emerging Applications of AI in Musculoskeletal Imaging

Speaker: Ali Guermazi, MD

Explores artificial intelligence applications in MSK radiology.

Relevant areas may include:

• Fracture detection
• Osteoarthritis
• Musculoskeletal MRI
• Quantitative imaging
• Segmentation
• Automated measurements
• Risk prediction

Module 5: AI Use Cases in Pediatric, Cardiothoracic, IR & Nuclear Medicine

Pediatric Radiology Artificial Intelligence

Speaker: Edward Lee, MD, MPH

Reviews the opportunities and unique challenges of AI in pediatric medical imaging.

Important considerations include:

• Pediatric datasets
• Age-specific imaging
• Radiation optimization
• Generalizability
• Workflow
• Clinical safety

Artificial Intelligence in Cardiothoracic Imaging

From Decision Support to Prognostic Biomarkers

Speaker: Fernando Kay, MD

Explores AI across thoracic and cardiovascular imaging.

Potential applications include:

• Chest imaging
• Cardiovascular imaging
• Disease detection
• Quantitative biomarkers
• Prognostic prediction
• Clinical decision support

Applications of Artificial Intelligence in Interventional Radiology

Speaker: Satvik Tripathi

AI may influence multiple stages of interventional radiology care.

Applications can include:

• Procedure planning
• Patient selection
• Image guidance
• Risk prediction
• Workflow
• Outcome prediction

Nuclear Medicine AI

Speaker: Babak Saboury, MD

Reviews artificial intelligence applications in nuclear medicine and molecular imaging.

Relevant areas may include:

• PET
• SPECT
• Reconstruction
• Segmentation
• Quantitative imaging
• Disease detection
• Prognostic analysis

Module 6: AI Research & Education

Demonstration of an AI Model Development Process From A to Z in Radiology

Speaker: Dooman Arefan, PhD

Provides insight into the complete AI-development workflow.

A typical model-development process may include:

Clinical Question → Dataset → Annotation → Model Development → Training → Validation → Testing → Clinical Evaluation

Understanding this process helps radiologists critically evaluate published AI research and commercial products.

Artificial Intelligence Research in Radiology: Team, Approach and Direction

Speaker: Shandong Wu, PhD

Successful AI research is inherently multidisciplinary.

Teams may include:

• Radiologists
• Data scientists
• Engineers
• Medical physicists
• Statisticians
• Computer scientists
• Clinical researchers

The session addresses practical approaches to conducting meaningful AI research in radiology.

Clinically Fluent, AI Literate: Teaching Radiologists About and With AI

Speaker: Justin Peacock, MD, PhD

Radiologists increasingly need AI literacy even if they never develop an algorithm.

Important educational areas include:

• Basic AI concepts
• Understanding model performance
• Recognizing limitations
• Evaluating research
• Understanding bias
• Safe clinical use
• Effective interaction with AI tools

AI can also become a tool for radiology education itself.

Module 7: Humanity & AI

The final module addresses one of the most important questions surrounding artificial intelligence:

What is the role of the radiologist in an increasingly AI-powered healthcare system?

Radiologist-Artificial Intelligence Collaboration and Teaming

Speaker: Florence Doo, MD

Instead of considering AI solely as automation, this session explores how radiologists and AI systems can work together.

Potential benefits of effective collaboration include:

• Increased efficiency
• Improved detection
• Quantitative support
• Reduced repetitive work
• Better prioritization
• More informed decision-making

Human oversight remains essential for interpreting AI output within the full clinical context.

Bias and Fairness of Artificial Intelligence in Radiology

Current State and Future Directions

Speaker: Judy Gichoya, MD, MS

AI algorithms can reproduce or amplify biases present in their training data.

Important considerations include:

• Dataset representation
• Population differences
• Demographic bias
• Algorithm performance across groups
• Generalizability
• Health disparities
• Fairness
• Responsible validation

Understanding these issues is essential for safe deployment.

Radiologists Fit in AI-Powered Radiology Services That Are Radiologist-Centered

Speaker: Eduardo Barbosa, MD, MBA

The course concludes by emphasizing that AI-powered radiology should remain centered around:

• Patient care
• Clinical judgment
• Radiologist expertise
• Communication
• Multidisciplinary collaboration
• Quality
• Safety

The objective is not AI-centered healthcare but radiologist-centered healthcare enhanced by AI.

Major AI Topics Covered

• Artificial Intelligence in Radiology
• Clinical AI
• Deep Learning
• Machine Learning
• Neural Networks
• Natural Language Processing
• NLP
• Large Language Models
• LLMs
• Generative AI
• Agentic AI
• Radiomics
• Foundation Models
• Multimodal Foundation Models
• AI Dataset Curation
• AI Model Development
• AI Validation
• AI Deployment
• AI Governance
• AI Regulation
• Medical AI Ethics
• AI Legal Considerations
• Algorithmic Bias
• AI Fairness
• Human-AI Collaboration
• AI Workflow Automation
• AI-Assisted Image Interpretation
• Clinical Decision Support
• AI Research
• AI Education

Radiology Subspecialties Covered

Clinical AI applications are reviewed across:

• Breast Imaging
• Neuroradiology
• Abdominal Imaging
• Musculoskeletal Imaging
• Pediatric Radiology
• Cardiothoracic Imaging
• Interventional Radiology
• Nuclear Medicine
• CT Imaging
• General Radiology

This broad subspecialty coverage makes the course useful even for radiologists who do not specialize specifically in imaging informatics.

AI Governance & Clinical Implementation

One of the strongest features of the program is its emphasis on moving beyond AI theory toward real-world clinical deployment.

AI implementation requires collaboration between:

• Radiologists
• Hospital leadership
• IT
• PACS teams
• Informatics specialists
• AI vendors
• Legal teams
• Compliance teams
• Quality and safety departments

The program provides a framework for considering the complete AI lifecycle rather than only algorithm accuracy.

AI, LLMs & Generative AI in Radiology

Large language models and generative AI represent one of the fastest-moving areas of healthcare technology.

Potential radiology applications include:

• Report generation
• Report summarization
• Clinical information extraction
• Decision support
• Patient communication
• Medical education
• Research
• Workflow automation
• Knowledge retrieval

However, clinicians must remain aware of limitations such as:

• Hallucination
• Inaccurate output
• Bias
• Privacy
• Data security
• Clinical validation
• Accountability

AI & Radiology Workflow

Artificial intelligence can influence radiology before, during, and after image interpretation.

Before Interpretation

AI may support:

• Protocoling
• Worklist prioritization
• Prior-study retrieval
• Scheduling

During Interpretation

AI may assist with:

• Detection
• Classification
• Segmentation
• Measurement
• Quantification
• Decision support

After Interpretation

AI may contribute to:

• Reporting
• Follow-up management
• Communication
• Quality assurance
• Research data extraction

Who Should Take This Course?

ARRS Clinical Artificial Intelligence in Radiology 2026 is particularly relevant for:

• Diagnostic Radiologists
• Radiology Residents
• Radiology Fellows
• Academic Radiologists
• Private Practice Radiologists
• Imaging Informatics Specialists
• Medical Imaging Professionals
• Radiology Researchers
• Medical Physicists
• Nuclear Medicine Physicians
• Interventional Radiologists
• Breast Radiologists
• Neuroradiologists
• Abdominal Radiologists
• Musculoskeletal Radiologists
• Pediatric Radiologists
• Cardiothoracic Radiologists
• Physicians involved in clinical AI implementation
• Healthcare professionals interested in medical artificial intelligence

Why This Course Is Useful

The main strength of ARRS Clinical Artificial Intelligence in Radiology 2026 is that it goes significantly beyond a basic introduction to AI.

The course progresses through:

AI Fundamentals → Clinical Implementation → Multimodal AI → Subspecialty Applications → Research → Education → Human-AI Collaboration

This provides a comprehensive framework for radiologists who want to:

• Understand modern AI terminology
• Learn the fundamentals of deep learning
• Understand LLMs and generative AI
• Explore agentic AI
• Understand multimodal foundation models
• Evaluate commercial AI products more critically
• Implement AI in clinical practice
• Understand AI governance
• Recognize regulatory and legal challenges
• Improve AI research literacy
• Explore AI applications in their subspecialty
• Understand bias and fairness
• Maintain a human-centered approach to AI

Product Information

• Product: ARRS Clinical Artificial Intelligence in Radiology 2026
• Provider: American Roentgen Ray Society – ARRS
• Year: 2026
• Original Course Dates: April 12–13, 2026
• Course Type: ARRS Categorical Course
• Modules: 7
• Specialty: Radiology / Artificial Intelligence
• Format: Digital Educational Content
• Included: Video Lectures + eBook
• eBook: 182 Pages / 24 Chapters
• Language: English

CME / Certificate Notice

The official ARRS enduring activity is designated for up to 20 AMA PRA Category 1 Credits™, subject to ARRS participation and credit-claiming requirements.

If this product consists of independent digital course files, purchase of those files should not be advertised as including official ARRS CME credit or an ARRS certificate unless access to those benefits is explicitly provided through ARRS.

4. Short Description

ARRS Clinical Artificial Intelligence in Radiology 2026 is a comprehensive AI-focused radiology program covering deep learning, NLP, LLMs, generative AI, agentic AI, radiomics, multimodal foundation models, AI deployment, governance, regulation, ethics, bias, research, education, and human-AI collaboration.

The course also reviews clinical AI applications across breast, neuro, abdominal, MSK, pediatric, cardiothoracic, interventional radiology, and nuclear medicine.

Included: Video Lectures + eBook
Modules: 7
Specialty: Radiology / Artificial Intelligence

---

Course Content

Module 1: Getting to Know AI

Overview of Radiology Artificial Intelligence: Latest Progress
Tessa Cook, MD, PhD

Primer on Artificial Intelligence (AI): Deep Learning, Natural Language Processing and Large Language Models, Generative AI, Agentic AI, and Radiomics
Hyun Soo Ko, MD

Artificial Intelligence Can Improve Radiology Workflow Efficiency By Automating Noninterpretive Tasks
Linda Moy, MD

Artificial Intelligence to Improve Radiology Imaging Interpretation
Shandong Wu, PhD

Module 2: AI Clinical Implementation

Legal and Ethical Considerations in AI Implementation
Julian Rivera, JD

Artificial Intelligence Deployment
Tessa Cook, MD, PhD

Artificial Intelligence (AI) Regulation and Governance: A Practice Perspective on How to Govern Assessment, Deployment, and Maintenance of AI Algorithms
Melissa Davis, MD, MBA

Panel Discussion
Linda Moy, MD (Moderator); Julian Rivera, JD; Tessa Cook, MD, PhD; Melissa Davis, MD, MBA

Module 3: Going Beyond Images to Multimodality

Medical Imaging Dataset Curation for Artificial Intelligence
Heather Whitney, PhD

Multimodal Foundation Models in Radiology
Christian Bluethgen, MD

Physics and Artificial Intelligence in CT
Lifeng Yu, PhD

Panel Discussion
Heather Whitney, PhD; Christian Bluethgen, MD; Lifeng Yu, PhD

Module 4: AI Use Cases in Subspecialties – Breast, Neuro, Abdominal & MSK

Breast Imaging Artificial Intelligence
Constance Lehman, MD, PhD

Artificial Intelligence in Neuroradiology
Paulo Kuriki, MD

Artificial Intelligence in Abdominal Imaging
Yee Ng, MD

Current and Emerging Applications of AI in Musculoskeletal Imaging
Ali Guermazi, MD

Module 5: AI Use Cases – Pediatrics, Cardiothoracic, IR & Nuclear Medicine

Pediatric Radiology Artificial Intelligence
Edward Lee, MD, MPH

Artificial Intelligence in Cardiothoracic Imaging: From Decision Support to Prognostic Biomarkers
Fernando Kay, MD

Applications of Artificial Intelligence in Interventional Radiology
Satvik Tripathi

Nuclear Medicine AI
Babak Saboury, MD

Module 6: AI Research and Education

Demonstration of an Artificial Intelligence Model Development Process (From A to Z) in Radiology
Dooman Arefan, PhD

Artificial Intelligence Research in Radiology: Team, Approach, and Direction
Shandong Wu, PhD

Clinically Fluent, AI Literate: Teaching Radiologists About and With AI
Justin Peacock, MD, PhD

Panel Discussion
Shandong Wu, PhD (Moderator); Dooman Arefan, PhD; Justin Peacock, MD, PhD

Module 7: Humanity and AI

Radiologist-Artificial Intelligence Collaboration and Teaming
Florence Doo, MD

Bias and Fairness of Artificial Intelligence in Radiology: Current State and Future Directions
Judy Gichoya, MD, MS

Radiologists Fit in AI-Powered Radiology Services That Are Radiologist-Centered
Eduardo Barbosa, MD, MBA

Panel Discussion
Charles Kahn, Jr., MD, MS (Moderator); Florence Doo, MD; Judy Gichoya, MD, MS; Eduardo Barbosa, MD, MBA