Pet-Based Radiomics Analysis for Predicting Prognosis and Differentiation Treatment-Related Changes in Glioma: A Systematic Review Publisher



Shakeri M ; Amraee A ; Hosseini S M ; Darvish L ; Farkhondeh F ; Mostaar A ; Ghadiri H
Source: Journal of Biomedical Physics and Engineering Published:2026

Abstract

Background: The assessment of treatment-induced changes in glioma and the evaluation of glioma prognosis are crucial components of effective treatment management. Radiomics models based on Positron Emission Tomography (PET) imaging can provide critical insights into therapeutic response monitoring. Objective: This systematic review aimed to evaluate the performance of PET-based radiomics models in distinguishing treatment-related changes and predicting the prognosis of glioma. Material and Methods: In this systematic review, the articles were searched from the Web of Science databases, MEDLINE, PubMed, and EMBASE. The search terms were “amino acid PET”, “PET”, “glioblastoma”, “glioma”, “positron emission tomography”, “machine learning”, “deep learning”, “radiomics”, “artificial intelligence”, “AI”, “prognosis”, “outcome”, “post treatment changes”, “treatment-related changes”, “progression”, “true progression” “pseudo-progression”, and “necrosis”. The titles, abstracts, and full text of the recognized citations were reviewed by two independent reviewers and then the selected articles were abstracted by two independent reviewers based on a standard grid. PRISMA checklist was applied to assess the overall quality of evidence for each outcome. Results: The PET-based radiomics models outperform conventional PET parameter models, such as maximum tumor-to-brain ratios and mean tumor-to-brain ratios in distinguishing post-treatment changes and predicting glioma prognosis. The model integrating radiomics features and the conventional PET parameters achieved superior diagnostic performance compared to radiomics and conventional parameter models solely in differentiation treatment related changes. Conclusion: PET based radiomics models demonstrate enhanced capability in differentiating tumor recurrence from treatment-related changes. The implementation of these models can facilitate personalized treatment plans and increase the patient’s overall survival or quality of life. © Journal of Biomedical Physics and Engineering.