Learning goals | ||
---|---|---|
1 | Follow fundamental architecture, structure, and implementation of AI systems to gain hands-on experience | This includes gaining experience on the following: |
a) The utility of AI in screening and triage | ||
b) The role of AI in precision medicine | ||
c) The limitations of the use AI in clinical practice | ||
2 | Understand how AI and machine learning approaches can be exploited for clinical research and associated shortcomings | These include recognition of the following: |
a) The significance of image quality and number of observations for effective AI application | ||
b) Consequences of using multi-institutional/multi-scanner data for research | ||
c) Overfitting, superfluous results, and their consequences | ||
3 | Broadly understand types of existing major open-source and commercial platforms that are already in use in clinical practice | This includes understanding and explain the following: |
a) The types and capabilities of commercially available AI systems and systems based on machine learning approaches | ||
b) How these can offer diagnostic support, such as image classification/detection, image segmentation, image registration, anomaly detection, and cross modality synthesis | ||
c) How they can integrate these systems into their clinical workflow to reduce burden on radiologists |