Random forest algorithm for precision dose prediction in brain cancer radiotherapy
ABSTRAK
Improving dose optimization during clinical planning using the treatment planning system for radiotherapy patients is crucial, yet executing this process can be time-consuming and reliant on the expertise of medical physicists. This research focuses on dose prediction employing machine learning for the planning target volume (PTV) and organ at risk (OAR) in cases of brain cancer treated with the volumetric modulated arc therapy planning technique. Utilizing DICOM planning data from brain cancer cases, this study utilizes extracted radiomic and dosiomic values as inputs and outputs for the research, employing a random forest algorithm model. Evaluation of the model reveals its effectiveness in predicting doses for PTV in brain cancer and OAR, with predicted homogeneity index and conformity index values of 0.14 ± 0.04 and 0.95 ± 0.01, respectively, compared to clinical values of 0.14 ± 0.13 and 0.94 ± 0.13. Thus, the random forest model demonstrates proficiency in predicting doses for brain cancer PTV and OAR, with an mean square error value of 0.017.
KATA KUNCI
- Mean square error; OAR; PTV; p-value; random forest
CARA MENGUTIP
Luqyana Adha Azwat, Prawito Prajitno, Dwi Seno Kuncoro Sihono, Dewa Ngurah Yudhi Persada (2026) Random forest algorithm for precision dose prediction in brain cancer radiotherapy. Komunikasi Fisika Indonesia, 21(2), 183-186 https://doi.org/10.31258/jkfi.21.2.183-186
Luqyana Adha Azwat, Prawito Prajitno, Dwi Seno Kuncoro Sihono, Dewa Ngurah Yudhi Persada "Random forest algorithm for precision dose prediction in brain cancer radiotherapy," Komunikasi Fisika Indonesia, vol. 21, no. 2, pp. 183-186, 2026. doi: 10.31258/jkfi.21.2.183-186.
@article{azwat2026,
author = {Luqyana Adha Azwat and Prawito Prajitno and Dwi Seno Kuncoro Sihono and Dewa Ngurah Yudhi Persada},
title = {Random forest algorithm for precision dose prediction in brain cancer radiotherapy},
journal = {Komunikasi Fisika Indonesia},
year = {2026},
volume = {21},
number = {2},
pages = {183-186},
doi = {10.31258/jkfi.21.2.183-186},
issn = {2579-521X},
}