Radiomic features of breast parenchyma: assessing differences between FOR PROCESSING and FOR PRESENTATION digital mammography

Abstract

Objective To assess the similarity and differences of radiomics features on full field digital mammography (FFDM) in FOR PROCESSING and FOR PRESENTATION data. Methods 165 consecutive women who underwent FFDM were included. Breasts have been segmented into “dense” and “non-dense” area using the software LIBRA. Segmentation of both FOR PROCESSING and FOR PRESENTATION images have been evaluated by Bland–Altman, Dice index and Cohen’s kappa analysis. 74 textural features were computed: 18 features of First Order (FO), 24 features of Gray Level Co-occurrence Matrix (GLCM), 16 features of Gray Level Run Length Matrix (GLRLM) and 16 features of Gray Level Size Zone Matrix (GLSZM). Paired Wilcoxon test, Spearman’s rank correlation, intraclass correlation and canonical correlation have been used. Bilateral symmetry and percent density (PD) were also evaluated. Results Segmentation from FOR PROCESSING and FOR PRESENTATION gave very different results. Bilateral symmetry was higher when evaluated on features computed using FOR PROCESSING images. All features showed a positive Spearman’s correlation coefficient and many FOR-PROCESSING features were moderately or strongly correlated to their corresponding FOR-PRESENTATION counterpart. As regards the correlation analysis between PD and textural features from FOR-PRESENTATION a moderate correlation was obtained only for Gray Level Non Uniformity from GLRLM both on “dense” and “non dense” area; as regards correlation between PD and features from FOR-PROCESSING a moderate correlation was observed only for Maximal Correlation Coefficient from GLCM both on “dense” and “non dense” area. Conclusions Texture features from FOR PROCESSING mammograms seem to be most suitable for assessing breast density.

Robustness of Radiomics in Pre-Surgical Computer Tomography of Non-Small-Cell Lung Cancer

“…The results of our study are in line with previous studies [ 3 , 4 , 13 , 15 , 23 , 25 , 26 , 30 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 ]; however, as a difference, we specifically addressed segmentation variability with a fixed CT device but analysed the dependence on bin-width, distance, resolution or interpolator.…”

“…However, they used only one segmentation obtained in consensus by three radiologists which might not be available in clinical routine. Many studies [ 1 , 2 , 3 , 4 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ] highlighted that from a clinical point of view, before radiomic features can be introduced into the routine clinical evaluation of non-small-cell lung cancer (NSCLC) patients, robustness with respect to segmentation variability must also be assessed [ 31 , 32 , 33 , 34 ].…”

An update on radiomics techniques in primary liver cancers

“…Images assessed during this analysis are accumulated from different centers or data-centers; so, these images could be obtained employing different manufacturers, with diverse protocols and parameters. These elements could influence radiomic models [ 61 , 62 ].…”

A Statistical Approach to Assess the Robustness of Radiomics Features in the Discrimination of Mammographic Lesions

“…Although several studies have investigated the repeatability of radiomics features, most of them focused on the following: a comparison across different manufacturers [50,51] and different image storage formats (raw vs. processed) [52] in full-field digital mammography (FFDM); a comparison between various AI-based feature extraction and segmentation approaches, e.g., radiomics and DL in magnetic resonance imaging (MRI) [53], manual and DL segmentation in in MRI [54,55], and computed tomography (CT) [56]; and the study of the impact of both clinical and radiomics features in contrast-enhanced spectral mammography (CESM) [57,58], in MG [59], and in FFDM [60]. Despite these valuable research efforts, and although the IBSI initiative addresses the issue of the reproducibility of radiomics characteristics, it is still difficult to interpret radiomics signatures and most studies have concentrated on repeatability using various imaging devices or reconstruction methods but employing one segmentation, which might not be accessible in clinical practice.…”