JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Development of a prognostic nomogram for ocular melanoma: a SEER population-based study (2000-2021)

  • 0Department of Ophthalmology, The First Hospital of Putian City, Putian, China.
Frontiers in Medicine +

|

February 13, 2025

|

February 13, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study developed a prognostic nomogram for ocular melanoma (OM) using machine learning. The nomogram accurately predicts survival, aiding clinical decisions for this rare cancer.

Area Of Science

  • Ophthalmology
  • Oncology
  • Medical Informatics

Background

  • Ocular melanoma (OM) is a rare but aggressive cancer with significant mortality.
  • Accurate prognostic tools are crucial for effective clinical decision-making and patient management in OM.

Purpose Of The Study

  • To develop and validate a prognostic nomogram for ocular melanoma (OM) to improve survival prediction.
  • To identify key independent prognostic variables for overall survival (OS) in OM patients.

Main Methods

  • Utilized univariate and multivariate COX proportional hazard regression to identify prognostic variables.
  • Developed a nomogram incorporating 13 significant clinicopathological factors.
  • Evaluated nomogram performance using ROC curves, calibration plots, DCA, and 10-fold cross-validation, comparing it with a machine learning model.

Main Results

  • Identified 13 independent prognostic factors for OS, including age, tumor characteristics, and metastasis.
  • The nomogram achieved a concordance index of 0.712 and AUCs of 0.749 (3-year), 0.734 (5-year), and 0.730 (10-year).
  • Calibration plots showed close agreement, DCA indicated superior net benefit, and the machine learning model achieved an AUC of 0.750.

Conclusions

  • A robust nomogram integrating 13 clinicopathological variables for OM prognosis was successfully developed.
  • The nomogram demonstrates strong predictive performance for survival outcomes, validated by multiple statistical methods and machine learning.
  • This tool can enhance prognosis prediction and support clinical decision-making for ocular melanoma patients.

Keywords: