Page 106 - ARNM-3-2

Advances in Radiotherapy &

Nuclear Medicine

LETTER TO EDITOR
Artificial intelligence in radiation oncology

Melek Yakar*
Department of Radiation Oncology, Faculty of Medicine, Osmangazi University, Eskişehir, Turkey

Dear Editor,
Radiation oncology plays a pivotal role in cancer treatment and the techniques and
instruments involved are constantly refined through technological advancements.
In recent years, artificial intelligence (AI) applications have garnered attention in this
domain due to their potential to enhance treatment processes, improve accuracy, and
personalize patient care. Particularly, the use of AI algorithms across a broad spectrum
– from radiotherapy (RT) planning stages to predicting oncological outcomes and
treatment-related toxicities—offers significant advantages to both clinicians and
patients. 1
AI applications in radiation oncology extend from clinical decision-making
algorithms to tumor and organ segmentation, and the evaluation of radiological images.
2
However, challenges such as data privacy, biases in algorithms, and difficulties in clinical
implementation highlight the need for further research in this area.

This letter discusses the present applications and future potentials of AI in radiation
oncology, emphasizing the technical and ethical challenges encountered during the
integration of these technologies.

The steps in radiation oncology include clinical evaluation of the patient, simulation,
segmentation, and fusion of imaging techniques, RT planning, quality assurance,
*Corresponding author:
Melek Yakar treatment delivery, and patient follow-up (oncological outcomes and treatment
3
(myakar@ogu.edu.tr) toxicities), with AI-related studies present at each step. The RT workflow begins with
patient evaluation. The clinicians must consider multiple parameters, such as the patient’s
Citation: Yakar M. Artificial
intelligence in radiation oncology. disease stage, histopathology, surgical reports (if available), medical imaging studies,
Adv Radiother Nucl Med. age, performance status, comorbidities, and hematological and biochemical parameters,
2025;3(2):98-101. to make a treatment decision.
doi: 10.36922/arnm.8429
Received: January 7, 2025 Although disease stage appears to be the most critical parameter when making
decisions guided by clinical guidelines, other factors such as radiomics and genetic
Accepted: March 14, 2025
mutations significantly influence disease prognosis. However, it is challenging for a
Published online: March 27, 2025 clinician to simultaneously evaluate and decide upon such a large dataset. Therefore,
Copyright: © 2025 Author(s). AI-supported decision systems are expected to play a critical role in personalized
This is an Open-Access article treatments.
distributed under the terms of the
Creative Commons Attribution AI also has the potential to automate, accelerate, and standardize the processes
License, permitting distribution, during segmentation and treatment planning. Deep learning (DL) algorithms assist
and reproduction in any medium, clinicians in segmenting both target volumes and organs at risk (OARs), speeding up
provided the original work is
properly cited. the processes while reducing inter-operator variability, resulting in more standardized
segmentation. 4
Publisher’s Note: AccScience
Publishing remains neutral with The processing and analysis of radiological images is another key area where
regard to jurisdictional claims in
published maps and institutional AI is effectively employed. Radiomics, a method used to extract advanced features
affiliations. from images, provides valuable insights into tumor biology, facilitating personalized

Volume 3 Issue 2 (2025) 98 doi: 10.36922/arnm.8429