Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{204707,
        author = {Annasreetha A C and Castro Iniyan P S and Gopika shree R and Ajay Kumar S R},
        title = {Longitudinal Validation of a Precision Prognostic Model for Uveal Melanoma},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {4079-4087},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=204707},
        abstract = {Uveal melanoma is known to be the most commonly occurring malignant intraocular tumor among adults and is associated with substantial clinical problems owing to its late-stage metastasis and poor prognosis. Prognosis plays a crucial role in managing patients, determining treatment plans, and devising a proper monitoring process. The majority of the current predictive systems only provide short-term validation processes and cannot establish their reliability over long clinical follow-up durations.
The present study offers a longitudinally validated prognostic framework that employs the machine learning method of the XGBoost algorithm via the Python programming language. This framework is expected to estimate personalized survival chances within the time windows of 1 year, 3 years, and 5 years using a variety of clinical, demographic, and tumor factors at the time of diagnosis. Longitudinal validation was performed through the methods of receiver operating characteristic analysis, calibration, Kaplan–Meier survival estimation, and temporal consistency. Results show high levels of discrimination power, calibration, and reliability. This study proves the usefulness of applying machine learning approaches in precision medicine and also emphasizes the significance of conducting longitudinal validation studies. In addition to this, the suggested model provides a robust and scalable method of survival prediction in uveal melanoma patients in the long run. The ability of the model to account for interdependencies between clinical covariates increases prognostic precision. The evaluation scheme for time series data provides more certainty regarding the temporal nature of the predictions made by the model.},
        keywords = {Uveal melanoma, longitudinal validation, machine learning, XGBoost, survival prediction, precision oncology, Kaplan–Meier analysis, predictive analytics.},
        month = {June},
        }