Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Aug 8.
doi: 10.1007/s00066-025-02449-1. Online ahead of print.

Non-invasive prediction of the secondary enucleation risk in uveal melanoma based on pretreatment CT and MRI prior to stereotactic radiotherapy

Yagiz Yedekci  1   2 , Hidetaka Arimura  3 , Yu Jin  4 , Melek Tugce Yilmaz  5 , Takumi Kodama  4 , Gokhan Ozyigit  5 , Gozde Yazici  5
Affiliations

Non-invasive prediction of the secondary enucleation risk in uveal melanoma based on pretreatment CT and MRI prior to stereotactic radiotherapy

Yagiz Yedekci et al. Strahlenther Onkol. .

Abstract

Purpose: The aim of this study was to develop a radiomic model to non-invasively predict the risk of secondary enucleation (SE) in patients with uveal melanoma (UM) prior to stereotactic radiotherapy using pretreatment computed tomography (CT) and magnetic resonance (MR) images.

Materials and methods: This retrospective study encompasses a cohort of 308 patients diagnosed with UM who underwent stereotactic radiosurgery (SRS) or fractionated stereotactic radiotherapy (FSRT) using the CyberKnife system (Accuray, Sunnyvale, CA, USA) between 2007 and 2018. Each patient received comprehensive ophthalmologic evaluations, including assessment of visual acuity, anterior segment examination, fundus examination, and ultrasonography. All patients were followed up for a minimum of 5 years. The cohort was composed of 65 patients who underwent SE (SE+) and 243 who did not (SE-). Radiomic features were extracted from pretreatment CT and MR images. To develop a robust predictive model, four different machine learning algorithms were evaluated using these features.

Results: The stacking model utilizing CT + MR radiomic features achieved the highest predictive performance, with an area under the curve (AUC) of 0.90, accuracy of 0.86, sensitivity of 0.81, and specificity of 0.90. The feature of robust mean absolute deviation derived from the Laplacian-of-Gaussian-filtered MR images was identified as the most significant predictor, demonstrating a statistically significant difference between SE+ and SE- cases (p = 0.005).

Conclusion: Radiomic analysis of pretreatment CT and MR images can non-invasively predict the risk of SE in UM patients undergoing SRS/FSRT. The combined CT + MR radiomic model may inform more personalized therapeutic decisions, thereby reducing unnecessary radiation exposure and potentially improving patient outcomes.

Keywords: Machine learning in medicine; Predictive model; Prognosis prediction; Radiomics; Side effect prediction.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: Y. Yedekci, H. Arimura, Y. Jin, M.T. Yilmaz, T. Kodama, G. Ozyigit, and G. Yazici declare that they have no competing interests.

Similar articles

See all similar articles

References

    1. Semeniuk O, Yu E, Rivard MJ (2024) Current and emerging radiotherapy options for uveal melanoma. Cancers 16(5)
    1. Reichstein DA, Brock AL (2021) Radiation therapy for uveal melanoma: a review of treatment methods available in 2021. Curr Opin Ophthalmol 32(3):183–190 - PubMed
    1. Kosydar S et al (2021) Systematic review and meta-analysis on the use of photon-based Stereotactic Radiosurgery versus Fractionated Stereotactic radiotherapy for the treatment of uveal melanoma. Am J Clin Oncol 44(1):32–42 - PubMed
    1. Bai H, Bosch JJ, Heindl LM (2023) Current management of uveal melanoma: a review. Clin Exp Ophthalmol 51(5):484–494 - PubMed
    1. van Beek JGM et al (2022) Fractionated stereotactic radiotherapy for uveal melanoma: long-term outcome and control rates. Acta Ophthalmol 100(5):511–519 - PubMed

LinkOut - more resources