Abstract
Radiation therapy (RT) remains a common treatment for many cancers but imparts considerable side effects that may persist for many years after treatment. In conventional RT, a single treatment plan is created based on the pre-treatment anatomy and delivered throughout several weeks of RT, despite tumor shrinkage and normal tissue anatomical changes that occur throughout RT. Adaptive RT (ART), in which one or more new treatment plans is created during a course of RT, has been increasingly employed in recent years to account for anatomical changes and spare normal tissues. In this chapter, we demonstrate the dosimetric impacts of ART using a head and neck cancer patient case and discuss two ART workflows: off-line ART with a conventional linac and on-line ART with an MR-linac. Presently, ART remains heavily operator-dependent and time-consuming, but the integration of machine learning into these workflows may help automate ART and increase its utilization.
Original language | English (US) |
---|---|
Title of host publication | Machine Learning and Artificial Intelligence in Radiation Oncology |
Subtitle of host publication | A Guide for Clinicians |
Publisher | Elsevier |
Pages | 365-380 |
Number of pages | 16 |
ISBN (Electronic) | 9780128220009 |
DOIs | |
State | Published - Jan 1 2023 |
Keywords
- Medical imaging
- Medical physics
- Oncology
- Radiation physics
- Radiology
ASJC Scopus subject areas
- General Agricultural and Biological Sciences
- General Biochemistry, Genetics and Molecular Biology