Artificial Intelligence Fundamentals Applications

This session builds upon the prior Artificial Intelligence fundamentals session and covers applications used in real world settings, including:

• Statistics versus Machine Learning
• Model Metrics
  • Classification Model Metrics – Accuracy, Precision, and Recall
  • Regression Model Metrics – MAE, MSE, RMSE, and R-squared
• Neural Networks and Their Applications
  • Feedforward Neural Networks
  • Increasing Complexity
  • Convolutional Neural Networks (CNN)
  • Recurrent Neural Network (RNN)
  • Transformer Learning
  • Generative Adversarial Networks (GAN)

This session will feature presentations by experts in data science at the Naval Information Warfare Systems Command (NAVWAR) 4.0 Data Science Learning Program.

FAIR and AI-Ready Data Sharing

This session is geared towards researchers wanting to understand and implement the skills and competencies needed to make biomedical and behavioral research data FAIR and AI-ready, including:

• FAIR principles and workflows
• Ethical considerations of data generating, handling, and sharing
• Core competencies to support AI-assisted biomedical research
• NIH-supported open-source data sharing
• Training and educational resources available to researchers

Different Privacy-Preserving Methodologies and Implications for Researchers

This session is geared towards researchers interested in learning about how privacy-preserving methodologies and technologies are used for data sharing and linkage to mitigate data privacy and the risk of re-identification of study participants, including:

• Importance of data privacy and participant confidentiality
• Privacy Preserving Record Linkage (PPRL)
• Differential privacy techniques
• Real-world applications and examples

Challenges, Opportunities, and Considerations for Researchers using Electronic Health Records and Real-World Data Sources

This session is geared towards researchers interested in leveraging real-world data in biomedical and behavioral research, including:

• What is Real-World Data (RWD)
• Scientific value and quality considerations for collection, use, and sharing of RWD
• Using RWD as part of the scientific paradigm
• RWD resources

Impact and Innovations from use of NIDDK-CR Resources

Hear from researchers who have accessed data from the NIDDK-CR on the impact and innovations from their research in the fields of data science, AI, and ML.