Impact of image preprocessing methods on reproducibility of radiomic features in multimodal magnetic resonance imaging in glioblastoma

Abstract

To investigate the effect of image preprocessing, in respect to intensity inhomogeneity correction and noise filtering, on the robustness and reproducibility of the radiomics features extracted from the Glioblastoma (GBM) tumor in multimodal MR images (mMRI).In this study, for each patient 1461 radiomics features were extracted from GBM subregions (i.e., edema, necrosis, enhancement, and tumor) of mMRI (i.e., FLAIR, T1, T1C, and T2) volumes for five preprocessing combinations (in total 116 880 radiomics features).The robustness and reproducibility of the radiomics features were assessed under four comparisons: (a) Baseline versus modified bias field; (b) Baseline versus modified bias field followed by noise filtering; (c) Baseline versus modified noise, and (d) Baseline versus modified noise followed bias field correction. The concordance correlation coefficient (CCC), dynamic range (DR), and interclass correlation coefficient (ICC) were used as metrics. Shape features and subsequently, local binary pattern (LBP) filtered images were highly stable and reproducible against bias field correction and noise filtering in all measurements. In all MRI modalities, necrosis regions (NC: n ̅~4 49/1461, 30%) had the highest number of highly robust features, with CCC and DR >= 0.9, in comparison with edema (ED: n ̅~2 96/1461, 20%), enhanced (EN: n ̅~2 81/1461, 19%) and active-tumor regions (TM: n ̅~2 54/1461, 17%). The necrosis regions (NC: n~449/1461, 30%) had a higher number of highly robust features (CCC and DR >= 0.9) than edema (ED: n~296/1461, 20%), enhanced (EN: n~281/1461, 19%) and active-tumor (TM: n~254/1461, 17%) regions across all modalities. Furthermore, our results identified that the percentage of high reproducible features with ICC >= 0.9 after bias field correction (23.2%), and bias field correction followed by noise filtering (22.4%) were higher in contrast with noise smoothing and also noise smoothing follow by bias correction. These preliminary findings imply that preprocessing sequences can also have a significant impact on the robustness and reproducibility of mMRI-based radiomics features and identification of generalizable and consistent preprocessing algorithms is a pivotal step before imposing radiomics biomarkers into the clinic for GBM patients.

Standardization of brain MR images across machines and protocols: bridging the gap for MRI-based radiomics

“…Radiomics relies on the extraction of features from multimodal imaging, aiming to improve patient care. Although acquisition parameters strongly affect the content of MR images, only some recent studies have specifically focused on the impact of MRI pre-processing methods on radiomics features 10,40,41 . Here, we investigated the impact of three different intensity normalization approaches combined with two grey-level discretization methods on brain MR-based radiomics.…”

“…Most recently, 2 publications focused on the image pre-processing steps and their impact on radiomics feature reproducibility in brain patients. Moradmand et al 10 evaluated the impact of 5 combinations of image pre-processing on the reproducibility of 1461 radiomic features (i.e., spatial resampling, skull stripping, noise reduction, bias field correction and intensity normalization) extracted from different glioblastoma (GBM) subregions (i.e., oedema, necrosis, enhanced tumour). They showed that radiomics features extracted from necrotic regions were the most reproducible and recommended that, after the bias field correction step, noise filtering should be applied.…”

“…The standard histogram is learned by averaging predefined landmarks deduced from histograms of the training set. The intensity landmark configuration C L = [1,10,20,30,40,50,60,70,80,90,99] (intensity percentiles) chosen in this study corresponds to the one defined by Shah et al 32 .…”

Machine Learning Based on a Multiparametric and Multiregional Radiomics Signature Predicts Radiotherapeutic Response in Patients with Glioblastoma

“…Image preprocessing was mainly performed through the FMRIB software library ( http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSL ) and with Python's SimpleITK package. To increase the robustness of features as much as possible through preprocessing [ 24 ], the following steps were adopted in this study: use of FLIRT in FMRIB to coregister the same T1WI image [ 25 ] as the template. After skull stripping, the isotropic voxel was resampled [ 26 ] to 1 × 1 × 1 mm 3 .…”

Development of a Radiomics Prediction Model for Histological Type Diagnosis in Solitary Pulmonary Nodules: The Combination of CT and FDG PET

“…In feature selection, we selected 2 CTR features, Skewness and Kurtosis (6) based on histogram, and 2 PETR features, SUVmean and SUVmax 7, with high reproducibility for slice thickness condition changes. The study of stability and reproducibility of the radiomics features (6,7,(24)(25)(26)(27)(28)(29)(30)(31) shows multiple parameter changes (e.g., slice thickness) in general produces greater measurement errors. In this case, the selected 4 features only have good reproducibility against slice thickness.…”