Radiomics Features-Based MRI In Glioblastoma Patients

Introduction: Glioblastoma multiforme (GBM) is the most common and aggressive primary malignant brain tumor and accounts for ۶۰% of brain tumors in adults. This tumor carries a poor prognosis, with a median survival of ۱۲–۱۵ months despite surgery followed by concurrent chemotherapy with temozolomide and radiation therapy(۱, ۲).In clinical routine, GBMs are usually diagnosed and followed-up with MRI. More recently, the field of radiomics has been introduced to extend the noninvasive study of oncologic tissue beyond established MR imaging metrics, and a large number of quantitative descriptors that reflect textural variations in image intensity, among other features, have been derived from imaging data. The aim of the new studies was to evaluate whether radiomic feature–based imaging signatures allow prediction of survival and stratification of patients with newly diagnosed glioblastoma(۳).Material and Methods: Scopus and Google Scholar databases were searched to find articles with relevant content for this presentation.The keywords used included Magnetic resonance imaging, Machine learning, Radiomics and Glioblastoma from ۲۰۱۲ to ۲۰۲۲.From the final results, articles were selected that provide a more relevant and complete explanation of the extraction of radiomics features-based MRI in glioblastoma patients.Results and Discussion: The results of various articles about MRI radiomics features in GBM patients were evaluated. In this presentation, we first introduce the characteristics of glioblastoma patients and then we examine the clinical applications of extracting radiomics features related to the MRI images of these patients.Conclusion: The aim of the new studies was to evaluate whether radiomic feature–based imaging signatures allow prediction of survival and stratification of patients with newly diagnosed glioblastoma.

Keywords:

Magnetic resonance imaging” , “Machine learning” , “Radiomics” , ”Glioblastoma