Peer-reviewedCC BYResearch
DEEP REINFORCEMENT LEARNING IN ONCOLOGY: ADVANCES IN CANCER IMAGING, RADIOTHERAPY, AND PERSONALIZED TREATMENT
- 1Consultant, MBBS, MD Radiodiagnosis, KIMS Hospital and Research Centre Krishna Rajendra Road, Parvathipuram, Vishweshwarapura, Basavanagudi, Bengaluru, Karnataka 560004
Abstract
Reinforcement Learning (RL) enables computers to learn dynamic decision making methods via trial and error, to maximize numerical reward signals. Its roots lie in early work across many fields, but recent advances have improved its capabilities greatly. Cancer is complex and dynamic and hence RL has a lot of potential in the field of oncology especially in imaging where it can improve accuracy and efficiency . In this study, we review RL in cancer and briefly introduce RL algorithms and categories. It then discusses applications of RL in cancer including in imaging and radiation. The report concludes with current challenges and future perspectives, emphasizing RL’s promise to personalize cancer diagnosis and therapy.
Related
- Counterfactual Policy Learning for Personalized Treatment Using SGLT2 Inhibitors
- Evolution of Precision Oncology, Personalized Medicine, and Molecular Tumor Boards
- Ross-PEARS: Autograft Reinforcement Using Personalized External Aortic Root Support
- Real-World Evidence as a Driver of Precision Oncology and Personalized Therapeutic Sequencing in Later-Line Metastatic Colorectal Cancer
- Precision Radiation Oncology: Molecular Strategies to Enhance Anti-Tumour Responses to Radiotherapy—Beyond Direct Tumour Cell Killing
BibTeX
@article{rajendran2026deep,
title = {{DEEP REINFORCEMENT LEARNING IN ONCOLOGY: ADVANCES IN CANCER IMAGING, RADIOTHERAPY, AND PERSONALIZED TREATMENT}},
author = {Ohmini Krishnamurthy Rajendran},
journal = {Research},
year = {2026},
month = jun,
doi = {10.5281/zenodo.20767336},
url = {https://doi.org/10.5281/zenodo.20767336},
note = {CC BY},
}
RIS
TY - JOUR
TI - DEEP REINFORCEMENT LEARNING IN ONCOLOGY: ADVANCES IN CANCER IMAGING, RADIOTHERAPY, AND PERSONALIZED TREATMENT
AU - Ohmini Krishnamurthy Rajendran
PY - 2026
DA - 2026/06/19
JO - Research
DO - 10.5281/zenodo.20767336
UR - https://doi.org/10.5281/zenodo.20767336
AB - Reinforcement Learning (RL) enables computers to learn dynamic decision making methods via trial and error, to maximize numerical reward signals. Its roots lie in early work across many fields, but recent advances have improved its capabilities greatly. Cancer is complex and dynamic and hence RL has a lot of potential in the field of oncology especially in imaging where it can improve accuracy and efficiency . In this study, we review RL in cancer and briefly introduce RL algorithms and categories. It then discusses applications of RL in cancer including in imaging and radiation. The report concludes with current challenges and future perspectives, emphasizing RL’s promise to personalize cancer diagnosis and therapy.
SN - 2407-9529
ER -