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

Abstract

Radiomics relies on the extraction of a wide variety of quantitative image-based features to provide decision support. Magnetic resonance imaging (MRI) contributes to the personalization of patient care but suffers from being highly dependent on acquisition and reconstruction parameters. Today, there are no guidelines regarding the optimal pre-processing of MR images in the context of radiomics, which is crucial for the generalization of published image-based signatures. This study aims to assess the impact of three different intensity normalization methods (Nyul, WhiteStripe, Z-Score) typically used in MRI together with two methods for intensity discretization (fixed bin size and fixed bin number). The impact of these methods was evaluated on first-and second-order radiomics features extracted from brain MRI, establishing a unified methodology for future radiomics studies. Two independent MRI datasets were used. The first one (DATASET1) included 20 institutional patients with WHO grade II and III gliomas who underwent post-contrast 3D axial T1-weighted (T1w-gd) and axial T2-weighted fluid attenuation inversion recovery (T2w-flair) sequences on two different MR devices (1.5 T and 3.0 T) with a 1-month delay. Jensen-Shannon divergence was used to compare pairs of intensity histograms before and after normalization. The stability of first-order and secondorder features across the two acquisitions was analysed using the concordance correlation coefficient and the intra-class correlation coefficient. The second dataset (DATASET2) was extracted from the public TCIA database and included 108 patients with WHO grade II and III gliomas and 135 patients with WHO grade IV glioblastomas. The impact of normalization and discretization methods was evaluated based on a tumour grade classification task (balanced accuracy measurement) using five well-established machine learning algorithms. Intensity normalization highly improved the robustness of first-order features and the performances of subsequent classification models. For the T1w-gd sequence, the mean balanced accuracy for tumour grade classification was increased from 0.67 (95% CI 0.61-0.73) to 0.82 (95% CI 0.79-0.84, P = .006), 0.79 (95% CI 0.76-0.82, P = .021) and 0.82 (95% CI 0.80-0.85, P = .005), respectively, using the Nyul, WhiteStripe and Z-Score normalization methods compared to no normalization. The relative discretization makes unnecessary the use of intensity normalization for the second-order radiomics features. Even if the bin number for the discretization had a small impact on classification performances, a good compromise was obtained

Robustness of magnetic resonance radiomic features to pixel size resampling and interpolation in patients with cervical cancer

“…To deal with nonstandardized MR intensity, the IN process could make MR radiomics more reliable. Several studies have demonstrated that the IN process was a necessary step for analyzing MR image features [ 22 , 23 , 34 ]. In contrast to a study by Carre et al, in which the IN process improved the robustness of first-order and second-order features [ 23 ], our study showed decreased robustness in IN images compared with non-IN images to pixel size resampling and interpolation (Figs.…”

“…Pyradiomics enabled the normalization of image intensity values. Normalization centered the image at the mean with standard deviation (SD) [ 22 , 23 ]. Normalization was based on all the gray values contained within the image and not just those defined by ROI.…”

Influence of Magnetic Field Strength on Magnetic Resonance Imaging Radiomics Features in Brain Imaging, an In Vitro and In Vivo Study

“…Recent articles using radiomics as biomarkers consider that the major challenge is that grayscale MRI intensities, contrary to X-ray CT, are not standardized and are highly dependent on manufacturer, sequence type and acquisition parameters ( 14 , 15 ).…”

Texture Analysis in Brain Tumor MR Imaging

“…Because many texture features are calculated based on histogram-style discretization, fixed bin size and fixed bin number (typically 32 or 64) are available options for discretization. 22 By applying generic image filters to the original images before feature extraction, the number of texture features can be easily doubled or tripled. Several current software programs offer a dozen kinds of these generic filters, and hundreds of features can be calculated depending on the parameter settings.…”