.png){kind=icon}
.png){kind=icon}
Return
Q-learning-driven real coded genetic algorithm for dynamic dose optimization in radiotherapy
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2026.115526.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Radiotherapy is modeled as a relaxed circle packing optimization problem. • A Q-Learning-based GA adapts crossover, mutation, and shot count dynamically. • QLRCGA achieves better tumor coverage and dose homogeneity than baseline methods. • QLRCGA maintains strong efficiency alongside improved optimization quality.
Keywords:
Radiotherapy
Q-Learning
Genetic Algorithm
Dose Optimization
Circle Packing
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W