Evaluating machine learning models and making decisions (7:06)

This video lesson will give you an understanding of:

• Why evaluating machine learning models is critical for safe decision-making in healthcare.

• Key performance metrics for binary classification, including accuracy, sensitivity, specificity, AUC, and calibration.

• The trade-offs between different metrics and how to evaluate regression models using MAE and MSE.

By the end of this video lesson, you will:

• Understand the difference between false positives and false negatives in clinical contexts.

• Be able to interpret common evaluation metrics to determine a model's reliability.

• Recognise that a good model must be interpretable, fair, and practical, not just accurate.

For a transcript of this video lesson, click ‘download handout’ above. Once you have completed the bite-sized video, please ensure you fill out our feedback survey which can be found at the end of the course page.