ToC of Course 3/5: Fundamentals of Machine Learning for Healthcare
#
Module 3: Concepts and Principles of Machine Learning in Healthcare
#
- Introduction to Deep Learning and Neural Networks
- Deep Learning and Neural Networks
- Cross Entropy Loss
- Gradient Descent
- Representing Unstructured Image and Text Data
- Convolutional Neural Networks
- Natural Language Processing and Recurrent Neural Networks
- The Transformer Architecture for Sequences
- Commonly Used and Advanced Neural Network Architectures
- Advanced Computer Vision Tasks and Wrap-Up
Module 4: Evaluation and Metrics for Machine Learning in Healthcare
#
- Introduction to Model Performance Evaluation
- Overfitting and Underfitting
- Strategies to Address Overfitting, Underfitting and Introduction to Regularization
- Statistical Approaches to Model Evaluation
- Receiver Operator and Precision Recall Curves as Evaluation Metrics
Module 5: Strategies and Challenges in Machine Learning in Healthcare
#
- Introduction to Common Clinical Machine Learning Challenges
- Utility of Causative Model Predictions
- Context in Clinical Machine Learning
- Intrinsic Interpretability
- Medical Data Challenges in Machine Learning Part 1
- Medical Data Challenges in Machine Learning Part 2
- How Much Data Do We Need?
- Retrospective Data in Medicine and “Shelf Life” for Data
- Medical Data: Quality vs Quantity
Module 6: Best Practices, Teams, and Launching Your Machine Learning Journey
#
- Clinical Utility and Output Action Pairing
- Taking Action - Utilizing the OAP Framework
- Building Multidisciplinary Teams for Clinical Machine Learning
- Governance, Ethics, and Best Practices
- On Being Human in the Era of Clinical Machine Learning
- Death by GPS and Other Lessons of Automation Bias
Module 7: Foundation Models
#
- Introduction to Foundation Models
- Adapting to Technology
- General AI and Emergent Behavior
- How Foundation Models Work
- Healthcare Use Cases for Text Data
- Healthcare Use Cases for Non-textual Unstructured Data
- Challenges and Pitfalls
- Conclusion