- Level Professional
- Duration 8 hours
- Course by University of Glasgow
-
Offered by
About
Machine learning systems used in Clinical Decision Support Systems (CDSS) require further external validation, calibration analysis, assessment of bias and fairness. In this course, the main concepts of machine learning evaluation adopted in CDSS will be explained. Furthermore, decision curve analysis along with human-centred CDSS that need to be explainable will be discussed. Finally, privacy concerns of deep learning models and potential adversarial attacks will be presented along with the vision for a new generation of explainable and privacy-preserved CDSS.Modules
Welcome to 'From machine learning models to clinical decision support systems'
1
Videos
- Welcome: From machine learning models to clinical decision support systems
A Guide to Model Validation in CDSS
1
Videos
- From Reproducibility to Generalisability
1
Readings
- An ABCD guide for prediction model validation in clinical settings
Calibration in Clinical Prediction Models
1
Videos
- A Guide to Model Validation in Clinical Decision Support Systems
1
Readings
- Calibration: the Achilles heel of predictive analytics
Calibration in Deep Learning Models
1
Videos
- Calibration of Deep Learning Models
1
Readings
- Bias assessment in Deep Learning Models
End of Week 1
1
Assignment
- End of week 1 Quiz
1
Discussions
- Week 1 - Your experience
Assessment of Risk of Bias in Clinical Studies
1
Videos
- Assessment of the Risk of Bias in EHR
1
Readings
- PROBAST: A Tool to Assess the Risk of Bias
'Unfairness' in state-of-the-art machine learning models
1
Videos
- Fairness in Machine Learning for Healthcare Applications (Part 1)
1
Readings
- Big Data's Disparate Impact
Strategies to ensure 'fairness' in machine learning models
1
Videos
- Fairness in Machine Learning for Healthcare Applications (Part 2)
1
Readings
- Ensuring Fairness in Machine Learning to Advance Health Equity
End of Week 2
1
Assignment
- End of week 2 Quiz
1
Discussions
- Week 2 - Your experience
Decision Curve Analysis
1
Videos
- Decision Curve Analysis
1
Readings
- A Guide to Interpreting Decision Curve Analysis
Ethics in Artificial Intelligence and Clinical Decision Support Systems
1
Videos
- Human-Centered Clinical Decision Support Systems
1
Readings
- A Roadmap Toward Transparent Expert Companions
Explainability as a central component of Human-Centered CDSS
1
Videos
- Evaluation of Explainability Models
1
Readings
- The role of explainability in creating trustworthy artificial intelligence for health care
End of week 3
1
Assignment
- End of week 3 Quiz
1
Discussions
- Week 3 - Your experience
Leakage and privacy concerns of deep learning models
1
Videos
- Privacy Concerns in CDSS
1
Readings
- Leakage and Privacy at Inference Time
Federated Learning and defences against privacy attacks
1
Videos
- Defences Against Inference Attacks
1
Readings
- Secure, privacy-preserving and federated machine learning
Adversarial attacks against explanations
1
Videos
- Adversarial Attacks - Explainability
1
Readings
- Explanations can be manipulated
End of week 4
1
Assignment
- End of week 4 Quiz
1
Discussions
- Week 4 - Your experience
Clinical Decision Support Systems
1
Assignment
- End of course summative quiz
Auto Summary
This professional-level course, offered by Coursera, delves into the evaluation of machine learning systems in Clinical Decision Support Systems (CDSS). It covers external validation, calibration analysis, bias assessment, and fairness. Learners will explore decision curve analysis, the need for explainable, human-centered CDSS, and address privacy concerns and adversarial attacks in deep learning models. The course spans 480 minutes and is available through Starter and Professional subscription plans, targeting professionals in Data Science and AI.

Fani Deligianni