Loading...

Course Description

This seven-week applied course introduces the foundations of artificial intelligence and predictive modeling in medicine using R. Students will learn to translate clinical questions into reproducible analyses using two complementary datasets: synthetic longitudinal health records generated with Synthea and real breast-cancer clinical data from METABRIC. The course progresses from exploratory data analysis, missing-data assessment, preprocessing, and feature engineering to logistic regression, individual patient prediction, ROC/AUC, classification thresholds, Kaplan-Meier analysis, Cox proportional hazards modeling, and validation with the concordance index. Emphasis is placed on clinical interpretation, evaluation on unseen patients, limitations of predictive models, and responsible communication rather than algorithmic complexity. By the end of the course, students will complete a reproducible clinical AI analysis and communicate its performance, clinical relevance, and limitations.

This course is ideal for NIH and biomedical research professionals, trainees, post-baccalaureate fellows, graduate students, clinicians, data analysts, and scientists who want practical experience applying predictive modeling to clinical data. It will support career development in biomedical data science, clinical research, translational research, precision medicine, and responsible use of AI by giving participants hands-on experience with reproducible R workflows, model evaluation, real-world cancer survival data, and clinically meaningful interpretation.

Learner Outcomes

When you complete the course successfully, you will be able to:
  • Prepare and explore clinical datasets in R, including missing-data assessment, categorical encoding, and feature engineering
  • Build and interpret logistic-regression models for individual and population-level clinical prediction
  • Evaluate classification models using ROC/AUC, sensitivity, specificity, precision, recall, and clinically meaningful thresholds
  • Analyze time-to-event cancer outcomes using Kaplan-Meier methods, Cox regression, hazard ratios, and the concordance index
  • Validate and communicate clinical prediction models using unseen test data while recognizing limitations, bias, and responsible-use considerations

Microcredential(s)

 

This course applies toward the Bioinformatics Endeavor and Artificial Intelligence and Machine Learning for Biomedical Researchers: Bioinformatics Specialization digital badges.

     

Textbook Information

There is no textbook for purchase required for this course.

Prerequisites

Basic familiarity with biomedical or clinical data is helpful but not required. Introductory knowledge of R or another programming/statistical language is recommended. The course is designed to reinforce the R skills needed for the analyses.

Additional Information

Required software: R (current stable release) and RStudio Desktop or Posit Cloud; free.
Required R packages will include tidyverse, tidymodels, survival, survminer, janitor, and related open-source packages.
Course datasets and instructor-provided R scripts will be supplied. No required textbook. Optional readings and software documentation will be provided.

Refund

Follow the link to review FAES Tuition Refund Policy.

Scholarship and Funding

Are you a self-funded student? FAES offers scholarship options. Click here for more information and to apply. 

Looking for resources to help you acquire funding for your continued education? Click here for our funding justification guide. 

Photo Release

By registering for this event, you agree to allow FAES to take photographs of you during the event and to use these photos for promotional purposes, including on our website, social media, and marketing materials, without further compensation. You understand that you have no right to review or approve the final use of these images.

Loading...

To Register Click on "Add to Cart"

Section Title
Predictive Modeling in Medicine Using R: Clinical Prediction and Survival Analysis
Type
Online Asynchronous
Dates
Mar 24, 2027 to May 11, 2027
Total Cost (Includes $100 non-refundable technology fee per course when applicable)
Eligible Discounts Can Be Applied at Checkout (2 Credits) $800.00
Available for Academic Credit
2 Credit(s)
Instructor(s)
Required fields are indicated by .