An Investigative Approach on the Prediction of Isocitrate Dehydrogenase (Idh1) Mutations and Co-Deletion of 1P19Q in Glioma Brain Tumors

Abstract

BACKGROUND - Isocitrate dehydrogenase (IDH)-wildtype, IDH-mutant and no 1p/19q codeletion, and IDH-mutant and 1p/19q-codeleted are the three categories of WHO grade II gliomas. There have been reports of prognostic and chemosensitivity variations among various molecular subtypes. For the molecular classification of lower grade gliomas, our goal was to test the predictive usefulness of imaging phenotypes evaluated. METHOD - IDH1 mutation and 1p/19q-codeletion scans of 253 patients with gliomas. Using the imaging characteristics’ prognostic accuracy for tumours of the IDH1-wild type was assessed. A validation set served as the model’s tested using Ensemble CNN Model and compare with VGG16 and VGG19 existing model also. RESULTS - Depending on the IDH1 mutation, distinct imaging characteristics were noticeably different. The Least Absolute Shrinkage and Selection Operator found that the nonlobar position, higher percentage of enhancing tumours, multifocal/multicentric distribution, and poorly defined non enhancing margins were all independent predictors of an IDH1 wild type. Comparing the IDH1-mutant, no codeletion group to the IDH1-mutant, 1p/19q-codeleted group, the latter typically exhibited mixed/restricted diffusion characteristics and demonstrated greater pial invasion. Using Ensemble CNN Classification got 95,95 and 96 accuracy, specificity and sensitivity respectively. CONCLUSIONS - According to the genetic markers of lower grade gliomas, preoperative MR imaging phenotypes vary, and they might be useful in determining the presence of the IDH1 mutation. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.