Repeatability of radiomic features in magnetic resonance imaging of glioblastoma: Test–retest and image registration analyses

Abstract

To assess the repeatability of radiomic features in magnetic resonance (MR) imaging of glioblastoma (GBM) tumors with respect to test-retest, different image registration approaches and inhomogeneity bias field correction. Methods: We analyzed MR images of 17 GBM patients including T1-and T2-weighted images (performed within the same imaging unit on two consecutive days). For image segmentation, we used a comprehensive segmentation approach including entire tumor, active area of tumor, necrotic regions in T1-weighted images, and edema regions in T2-weighted images (test studies only; registration to retest studies is discussed next). Analysis included N3, N4 as well as no bias correction performed on raw MR images. We evaluated 20 image registration approaches, generated by cross-combination of four transformation and five cost function methods. In total, 714 images (17 patients × 2 images × ((4 transformations × 5 cost functions) + 1 test image) and 2856 segmentations (714 images × 4 segmentations) were prepared for feature extraction. Various radiomic features were extracted, including the use of preprocessing filters, specifically wavelet (WAV) and Laplacian of Gaussian (LOG), as well as discretization into fixed bin width and fixed bin count (16, 32, 64, 128, and 256), Exponential, Gradient, Logarithm, Square and Square Root scales. Intraclass correlation coefficients (ICC) were calculated to assess the repeatability of MRI radiomic features (high repeatability defined as ICC ≥ 95%). Results: In our ICC results, we observed high repeatability (ICC ≥ 95%) with respect to image preprocessing, different image registration algorithms, and test-retest analysis, for example: RLNU and GLNU from GLRLM, GLNU and DNU from GLDM, Coarseness and Busyness from NGTDM, GLNU and ZP from GLSZM, and Energy and RMS from first order. Highest fraction (percent) of repeatable features was observed, among registration techniques, for the method Full Affine transformation with 12 degrees of freedom using Mutual Information cost function (mean 32.4%), and

The Pursuit of Generalizability to Enable Clinical Translation of Radiomics

“…Additionally, the findings suggested that both intensity and texture features demonstrated higher variability as compared with the shape features. These findings are in line with previous repeatability (5) and reproducibility studies (10,11), where intensity and texture features have been found to be sensitive to subtle variations in test-retest scans, as well as image variations across sites and scanners. Both texture (describing heterogeneity measures from gray-level co-occurrence matrix) and intensity features (describing the distribution of intensity values) are dependent on the actual per-voxel intensity values within the region of interest and may yield substantially different measurements if they are not appropriately normalized in some fashion before downstream analysis.…”

“…Multiple efforts are currently underway to address some of the practical challenges associated with generalizability of radiomic features. These include attempts to study radiomic variability across large multisite data (4), controlled test-retest data (5), and phantom studies (6). Most of these attempts so far have been in the pursuit of identifying a set of radiomic features that are both repeatable and reproducible across different image acquisition, preprocessing, segmentation, and radiomic feature extraction pipelines.…”

Prognostic and Predictive Value of Integrated Qualitative and Quantitative Magnetic Resonance Imaging Analysis in Glioblastoma

“…For MRI radiomics, such a unit does not exist which poses problems due to inter-and intra-scanner variability. Multiple pre-processing methods have been developed, though not all radiomics features were shown to be robust between different pre-processing approaches [57][58][59]. This calls for a generalized pre-processing pipeline and focus on features that are shown to be robust.…”

Title: Cardiac MRI derived radiomics based machine learning (ML) models in identifying pulmonary hypertension (PH)- Impact of segmentation, radiomics side experiments and DAFIT approach on model performance

“…For each mask, 348 features were extracted using the MaZda software. These included histogram (9), cooccurrence matrix (220), run-length matrix (20), gradient (5), autoregressive (5), geometrical (73) and wavelet (16) features. Details about these features are provided elsewhere and beyond the scope of this work.…”