An official website of the United States government 
Activity Outline
FDA Grand Rounds: Synthetic Data For Medical Imaging AI
November 14, 2024
Zoom Platform
FDA Grand Rounds: Synthetic Data For Medical Imaging AI
November 14, 2024
Zoom Platform
Activity Coordinators:
Isaac Miller (Isaac.Miller@fda.hhs.gov), Rokhsareh Shahidzadeh (Rokhsareh.Shahidzadeh@fda.hhs.gov)
Series Description
The FDA Grand Rounds highlights cutting-edge research underway across the agency and its impact on protecting and advancing public health. Each session features an FDA scientist presenting on a key public health challenge and how FDA is applying science to its regulatory activities.
Lecture Description
Artificial Intelligence (AI)-enabled medical imaging devices require access to large-scale and representative datasets for both training and evaluation. Obtaining sufficient data remains a crucial challenge for most applications in medical image analysis, in part due to patient privacy concerns, acquisition and annotation difficulties or high costs, limiting wider development of medical AI.
We will show that synthetic data, i.e., artificial data designed to approximate properties and relationships seen in patient data, can be used to supplement patient data in medical AI development and evaluation, mitigating data availability concerns. We will summarize and compare different methodologies for creating medical synthetic data, distinguishing between knowledge-based (KB) (e.g., mechanistic) and imaging-based (e.g., generative AI) models. We will demonstrate how KB synthetic data generation in breast and skin imaging applications can be effectively used for AI development and analysis.
Artificial Intelligence (AI)-enabled medical imaging devices require access to large-scale and representative datasets for both training and evaluation. Obtaining sufficient data remains a crucial challenge for most applications in medical image analysis, in part due to patient privacy concerns, acquisition and annotation difficulties or high costs, limiting wider development of medical AI.
We will show that synthetic data, i.e., artificial data designed to approximate properties and relationships seen in patient data, can be used to supplement patient data in medical AI development and evaluation, mitigating data availability concerns. We will summarize and compare different methodologies for creating medical synthetic data, distinguishing between knowledge-based (KB) (e.g., mechanistic) and imaging-based (e.g., generative AI) models. We will demonstrate how KB synthetic data generation in breast and skin imaging applications can be effectively used for AI development and analysis.
References
- A. Kim, N. Saharkhiz, E. Sizikova, M. Lago, B. Sahiner, J. G. Delfino, A. Badano,” S-SYNTH: Knowledge-Based, Synthetic Generation of Skin Images”. International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2024.
- E. Sizikova, N. Saharkhiz, D. Sharma, M. Lago, B. Sahiner, J. G. Delfino, A. Badano. Knowledge-based in silico models and dataset for the comparative evaluation of mammography AI for a range of breast characteristics, lesion conspicuities and doses. Conference on Neural Information Processing Systems (NeurIPS) 2023.
- A. Badano, M. A. Lago, E. Sizikova, J. G. Delfino, S. Guan, M. A. Anastasio, B. Sahiner. The stochastic digital human is now enrolling for in silico imaging trials—methods and tools for generating digital cohorts. Progress in Biomedical Engineering, 2023.
Series Objectives
- Discuss the research conducted at the FDA
- Explain how FDA science impacts public health
Learning Objectives After completion of this activity, the participant will be able to:
- Analyze different classes of techniques for generating synthetic medical imaging data and recognize their strengths and weaknesses.
- Identify how synthetic data can be used in various stages of the AI lifecycle.
Target Audience
This activity is intended for physicians, pharmacists, nurses, and other scientists within the agency and external scientific communities.
This activity is intended for physicians, pharmacists, nurses, and other scientists within the agency and external scientific communities.
Registration Information
Registration is complimentary; therefore refunds are not applicable. For information on how to register to attend this activity, please contact the Activity Coordinator(s) listed above.
Registration is complimentary; therefore refunds are not applicable. For information on how to register to attend this activity, please contact the Activity Coordinator(s) listed above.
Agenda
Lecture 1 November 14, 2024
Lecture 1 November 14, 2024
| Time | Topic | Speaker |
|---|---|---|
| 12:00 - 1:00 PM EST | Synthetic Data For Medical Imaging AI | Elena Sizikova, PhD |
Continuing Education Accreditation
In support of improving patient care, FDA Center for Drug Evaluation and Research is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC) to provide continuing education for the healthcare team.
This activity was planned by and for the healthcare team, and learners will receive 1 Interprofessional Continuing Education (IPCE) credit(s) for learning and change.
CME
FDA Center for Drug Evaluation and Research designates this live activity for a maximum of 1.00 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
CPE
This knowledge-based activity has been assigned ACPE Universal Activity Number JA0002895-0000-24-017-L99-P for 1.00 contact hour(s).
CNE
FDA Center for Drug Evaluation and Research designates this activity for 1.00 contact hour(s).
Requirements for Receiving CE Credit
All learners claiming credit must attest to their attendance and complete all required activity evaluation(s) in the FDA CE Portal (ceportal.fda.gov) within 14 days after an activity ends. Upon completion, learners may view/print statement of credit.
Attention NABP Pharmacists and Pharmacy Technicians: The FDA CE Team will report your credit to the National Association of Boards of Pharmacy (NABP) provided you add your NABP ID and date of birth to your profile in the FDA CE Portal. The only official Statement of Credit is the one you pull from CPE Monitor®. If you do not see your credit reflected on CPE Monitor®* after 45 days of attestation, please contact FDACETeam@fda.hhs.gov.
*CPE Monitor® sets a strict 60-day limit on uploading credits.
Disclosure
Faculty
Planning Committee
CE Consultation and Accreditation Team
Faculty
- Sizikova, Elena, PhD, Staff Fellow, DIDSR - nothing to disclose
Planning Committee
- Dinatale, Miriam, DO, Team Leader, Food and Drug Administration - nothing to disclose
- Pfundt, Tiffany, PharmD, Program Coordinator, FDA/CDER/OTBB - nothing to disclose
- Shahidzadeh, Rokhsareh, RN, MSN, Senior Regulatory Health Education Specialist, FDA - nothing to disclose
CE Consultation and Accreditation Team
- Faberlle, Alexandra M., Training Specialist / FDA/CDER/OEP/DLOD - nothing to disclose
- Bryant, Traci, M.A.T., Lead Training Specialist, FDA/CDER/OEP/DLOD - nothing to disclose
- Wood, Sara, Accreditation Program Administrator, CECAT, FDA/CDER/OEP/DLOD - nothing to disclose
Any relationship shown above in italics has been divested within the last 24 months and is therefore considered mitigated.
All relevant financial relationships have been mitigated.
All relevant financial relationships have been mitigated.
U.S. Food and Drug Administration