Automated feature learning and survival prognostication in grade 4 glioma using supervised machine learning models.

Journal: Journal of neuro-oncology
Published Date: Jun 16, 2025

Abstract

OBJECTIVE: WHO grade 4 glioma is the most common primary malignant brain tumor, with a median survival of only 14.6 months. Predicting survival outcomes remains challenging due to the tumor's heterogeneity and the influence of multiple clinical factors. Machine learning (ML) techniques have demonstrated superior predictive performance compared to traditional statistical models. Embedded feature-selection techniques such as Lasso shrinkage or Random-Forest importance scores are widely used, yet grade-4-glioma prognostic models still rely on an initial clinician-curated variable list and on ad-hoc cut-offs (e.g., "top X features" or "above certain threshold") when deciding how many ranked features to keep-choices that markedly influence model accuracy. We therefore developed a fully data-driven pipeline that begins with an unrestricted pool of clinical, functional, and biomarker variables, employs SHAP values for global importance ranking, and uses automated feature-subset optimization to identify the most optimal combination of predictors that maximizes survival-prediction performance in grade-4 glioma.

Authors

  • Yuncong Mao
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Linda Tang
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Melanie Alfonzo Horowitz
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Saket Myneni
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Jacob Gould
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Sai Chandan Reddy
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Kritika Gowda
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Michelle Rodriguez
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Anthony Rivetti
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Evan Li
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Ruiwen Xiong
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Rohan Venkatdas
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Max Saint-Germain
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Briana Santo
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Julian Gendreau
    5Department of Biomedical Engineering, Johns Hopkins Whiting School of Engineering, Baltimore, Maryland.
  • A Karim Ahmed
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Jordina Rincon-Torroella
    Department of Neurosurgery, Johns Hopkins University School of Medicine, 1800 Orleans Street, Baltimore, 21287, MD, USA.
  • Christopher Jackson
    College of Medicine, University of Tennessee Health Sciences Center, Memphis, TN, United States.
  • Gary Gallia
    Department of Neurosurgery, Johns Hopkins Medicine, Baltimore, MD, USA.
  • Chetan Bettegowda
    Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland.
  • Jon Weingart
  • Debraj Mukherjee
    Department of Neurosurgery, Johns Hopkins University School of Medicine, 1800 Orleans Street, Baltimore, 21287, MD, USA.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals