Differentiating solitary brain metastases from glioblastoma by radiomics features derived from MRI and 18F-FDG-PET and the combined application of multiple models

Abstract

This study aimed to explore the ability of radiomics derived from both MRI and 18F-fluorodeoxyglucose positron emission tomography (18F-FDG-PET) images to differentiate glioblastoma (GBM) from solitary brain metastases (SBM) and to investigate the combined application of multiple models. The imaging data of 100 patients with brain tumours (50 GBMs and 50 SBMs) were retrospectively analysed. Three model sets were built on MRI, 18F-FDG-PET, and MRI combined with 18F-FDG-PET using five feature selection methods and five classification algorithms. The model set with the highest average AUC value was selected, in which some models were selected and divided into Groups A, B, and C. Individual and joint voting predictions were performed in each group for the entire data. The model set based on MRI combined with 18F-FDG-PET had the highest average AUC compared with isolated MRI or 18F-FDG-PET. Joint voting prediction showed better performance than the individual prediction when all models reached an agreement. In conclusion, radiomics derived from MRI and 18F-FDG-PET could help differentiate GBM from SBM preoperatively. The combined application of multiple models can provide greater benefits.

Differentiating Glioblastoma Multiforme from Brain Metastases Using Multidimensional Radiomics Features Derived from MRI and Multiple Machine Learning Models

“…A number of previous studies have used multiparametric MRI data to discriminate between GBM tumors and MET, including advanced imaging methods such as diffusion, perfusion, and MR spectroscopy [1][2][3][4][5][6]. It should be noted that advanced imaging is not incorporated into all MRI protocols across all sites and is highly dependent on the acquisition and analysis method [14,34,41]. Therefore, it is important to be able to classify brain tumors based on common sequences [14,34,41].…”

“…[7][8][9][10][11][12]. In each slice, ROIs were drawn along the tumor margin to encompass the entire tumor area [14,17,[22][23][24]. The preprocessing and feature extraction of the images were performed using Pyradiomics (http://pyradiomics.readthedocs.io/en/latest/ index.html) [20,[24][25][26][27][28].…”

“…Following training, the machine can accurately identify the tumor type in a new sample and support clinical decisions significantly. In MRI images, brain tumors can be categorized in many ways [10][11][12][13][14]. One of the most prominent ones is fuzzy clustering means (FCM), support vector machine (SVM), artificial neural network (ANN), knowledge-based techniques, and the expectation-maximization (EM) algorithm methodology [14,[17][18][19].…”

“…Even though the two lesions had different treatment strategies, similar radiological appearances made it difficult to differentiate them. Many years of radiological research have focused on accurately distinguishing these two lesions [ 11 , 13 , 14 ].…”

A Survey of Radiomics in Precision Diagnosis and Treatment of Adult Gliomas

“…Sartoretti et al [15] extracted radiomics features from amide proton transfer-weighted imaging (APTWI) of 21 patients with gliomas and 27 patients with BM and established a prediction model based on a multi-layer perception algorithm for distinguishing gliomas from BM, achieving an AUC of 0.836. The study by Marginean et al [16] retrospectively analyzed contrast-enhanced CT (CECT) images of 36 patients with solitary brain tumors (17 HGGs and 19 BM) and used MaZda software (version 5) for the texture analysis of peritumoral edema to discriminate HGGs from BM, and the optimal texture parameter Perc10 had a sensitivity of 81.0%, specificity of 85.7%, and AUC of 0.84. Although previous studies have achieved good diagnostic performance, there is still much room for improvement in sensitivity and specificity.…”