Abstract
Assessing and controlling the quality of medical data such as images, as well as AI-derived parameters from these data, is an important component of clinical imaging and retrospective population studies. This chapter deals with the issue of how AI can be used to automatically control the quality of its results. The clinical introduction discusses the role of current role and potential of quality control techniques. The technical review summarizes the state-of-the-art in AI-enabled quality control. The main focus is on cardiac MR imaging, but applications in echocardiography are also introduced. Methods to identify motion artefacts, poor planning, missing slices and failed segmentations are discussed. The practical tutorial involves implementing a simple motion artefact detection model. The closing clinical opinion piece discusses the future role of AI in ensuring that the quality of images and their derived measures is maximized.
| Original language | English |
|---|---|
| Title of host publication | AI and Big Data in Cardiology |
| Subtitle of host publication | a Practical Guide |
| Publisher | Springer International Publishing |
| Pages | 135-156 |
| Number of pages | 22 |
| ISBN (Electronic) | 9783031050718 |
| ISBN (Print) | 9783031050701 |
| DOIs | |
| Publication status | Published - 1 Jan 2023 |
Bibliographical note
Publisher Copyright:© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
Keywords
- Artefact
- Artificial intelligence
- Deep learning
- Echocardiography
- Machine learning
- Magnetic resonance
- Motion
- Planning
- Quality control