JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Radiomic Features as Artificial Intelligence Prognostic Models in Glioblastoma: A Systematic Review and Meta-Analysis

  • 0Neurosurgery Division, Department of Surgery, Faculty of Medicine, Universitas Udayana, Udayana University Hospital, Denpasar 80361, Indonesia.
Diagnostics (basel, Switzerland) +

|

November 9, 2024

|

November 9, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Radiomics features (RFs) combined with artificial intelligence (AI) show significant prognostic value for glioblastoma patients. This approach aids in predicting overall survival (OS) and progression-free survival (PFS) noninvasively.

Area Of Science

  • Neuro-oncology
  • Medical Imaging Analysis
  • Artificial Intelligence in Medicine

Background

  • Glioblastoma is the most common primary brain tumor, accounting for 80% of central nervous system tumors.
  • Accurate prognosis is crucial for neuro-oncology, involving the assessment of disease progression over time.
  • Radiomic features (RFs) derived from MRI, utilizing artificial intelligence (AI), offer a noninvasive method for prognostic evaluation.

Purpose Of The Study

  • To systematically review and evaluate the prognostic significance of radiomic features (RFs) in glioblastoma.
  • To assess the utility of RFs combined with AI in predicting patient survival outcomes.

Main Methods

  • An extensive literature search was conducted across PubMed, ScienceDirect, EMBASE, Web of Science, and Cochrane databases up to July 25, 2024.
  • Included studies focused on glioblastoma, MRI, radiomics, and prognosis/survival in human subjects, excluding case reports and reviews.
  • Study quality was assessed using the Newcastle-Ottawa Scale (NOS), and Hazard Ratios (HRs) for Overall Survival (OS) and Progression-Free Survival (PFS) were pooled using random-effects models.

Main Results

  • Out of 253 initial studies, 14 met the eligibility criteria, including 12 for OS and 8 for PFS, with a total of 1,639 and 747 patients, respectively.
  • The pooled Hazard Ratio (HR) for Overall Survival (OS) was 3.59 (95% CI, 1.80-7.17).
  • The pooled Hazard Ratio (HR) for Progression-Free Survival (PFS) was 4.20 (95% CI, 1.02-17.32).

Conclusions

  • The integration of radiomic features (RFs) with artificial intelligence (AI) demonstrates significant prognostic value in glioblastoma.
  • This AI-driven radiomics approach provides a robust tool for predicting both overall survival (OS) and progression-free survival (PFS) in glioblastoma patients.

Keywords: