Robustness of Radiomic Features: Two-Dimensional versus Three-Dimensional MRI-Based Feature Reproducibility in Lipomatous Soft-Tissue Tumors

Abstract

This retrospective study aimed to compare the intra- and inter-observer manual-segmentation variability in the feature reproducibility between two-dimensional (2D) and three-dimensional (3D) magnetic-resonance imaging (MRI)-based radiomic features. The study included patients with lipomatous soft-tissue tumors that were diagnosed with histopathology and underwent MRI scans. Tumor segmentation based on the 2D and 3D MRI images was performed by two observers to assess the intra- and inter-observer variability. In both the 2D and the 3D segmentations, the radiomic features were extracted from the normalized images. Regarding the stability of the features, the intraclass correlation coefficient (ICC) was used to evaluate the intra- and inter-observer segmentation variability. Features with ICC > 0.75 were considered reproducible. The degree of feature robustness was classified as low, moderate, or high. Additionally, we compared the efficacy of 2D and 3D contour-focused segmentation in terms of the effects of the stable feature rate, sensitivity, specificity, and diagnostic accuracy of machine learning on the reproducible features. In total, 93 and 107 features were extracted from the 2D and 3D images, respectively. Only 35 features from the 2D images and 63 features from the 3D images were reproducible. The stable feature rate for the 3D segmentation was more significant than for the 2D segmentation (58.9% vs. 37.6%, p = 0.002). The majority of the features for the 3D segmentation had moderate-to-high robustness, while 40.9% of the features for the 2D segmentation had low robustness. The diagnostic accuracy of the machine-learning model for the 2D segmentation was close to that for the 3D segmentation (88% vs. 90%). In both the 2D and the 3D segmentation, the specificity values were equal to 100%. However, the sensitivity for the 2D segmentation was lower than for the 3D segmentation (75% vs. 83%). For the 2D + 3D radiomic features, the model achieved a diagnostic accuracy of 87% (sensitivity, 100%, and specificity, 80%). Both 2D and 3D MRI-based radiomic features of lipomatous soft-tissue tumors are reproducible. With a higher stable feature rate, 3D contour-focused segmentation should be selected for the feature-extraction process.

Tumor-to-bone distance and radiomic features on MRI distinguish intramuscular lipomas from well-differentiated liposarcomas

“…Tumor region of interests (ROIs) were drawn on the entire volume of the lesion. Additionally, a reference region of interest (ROI) was drawn in fat on T1W MRI for image-intensity-normalization procedure [ 24 ]. All cases were selected randomly and blindly for repetitive the segmented ROI by two observers (statistician and research scientist) and subsequently confirmed their precision by experts in musculoskeletal radiology and orthopedic oncology.…”

“…The purpose of this step was to adjust for differences in T1W MRI protocols. The normalized intensity value (NIV) [ 24 ] was determined as follows: .…”

“…Interpretation of ICC value is as follows: poor (ICC < 0.50); moderate (0.50 ≤ ICC < 0.75); good (0.75 ≤ ICC < 0.90); and excellent (ICC ≥ 0.90) [ 30 ]. In this study, we considered a feature reproducible if the ICC value was more than 0.75 [ 24 , 31 – 35 ]. Additionally, Pearson correlation coefficient ( r )/Spearman’s rank-order coefficient ( ) were also computed.…”

Radiomics in Musculoskeletal Tumors

“…Whether 3D approaches generate more consistent radiomics results is still debated, with some publications in sarcoma imaging asserting superiority of 3D segmentation versus 2D segmentation techniques. [24][25][26][27] Image segmentation can be performed with manual delineation of a lesion, semiautomatic segmentation approaches with reduced need for direct user interaction, or with completely automated segmentation procedures. Fully automated image segmentation in sarcoma is still an area of active research with reliable validated automatic segmentation techniques holding the promise of increased speed and reproducibility of tumor segmentations without the need for time-intensive manual or semiautomated segmentation techniques or intra-and interrater variability assessments.…”

Relaxed Adaptive Lasso for Classification on High-Dimensional Sparse Data with Multicollinearity

“…Particularly needed are predictive modeling techniques that can provide accurate results to help decision-making. Logistic regression is one of the techniques that is widely employed in data analysis and machine learning communities [1][2][3][4]. This predictive modeling technique describes the relationships between independent and outcome variables and predicts the outcome variables' future values [5,6].…”