This is the authors' abstract. We don't add key points for this paper.
To investigate the potential of multi- b -value diffusion-weighted imaging (DWI)–derived radiomics models for evaluating angiogenesis and spatial heterogeneity in orthotopic hepatocellular carcinoma (HCC) xenograft nude mouse models. Sixty-one HCC xenograft nude mice were randomly allocated into training and testing cohorts (8:2 ratio). Multi- b -value DWI was performed on a 1.5T MR scanner to obtain parametric maps of 4 diffusion models: intravoxel incoherent motion (IVIM), diffusion kurtosis imaging (DKI), stretched exponential model (SEM), and fractional-order calculus (FROC) model. The histopathological evaluation included microvessel density (MVD) quantification via anti-CD31 staining and tumor spatial heterogeneity assessment using the standard deviation (SD) of grayscale pixel distributions via hematoxylin-eosin (HE) staining. Radiomics features were extracted from DWI maps, and features were selected using logistic regression (LR), random forest (RF), and support vector machine (SVM) algorithms. Two radiomics models were constructed to discriminate between high and low MVD groups and SD groups. Model performance was evaluated using the areas under the curve (AUCs). Radiomics models were constructed with 15 principal components of selected features derived from multi- b -value DWI maps. In the testing cohort, the AUCs of the radiomics models for distinguishing high and low MVD and SD groups of HCC were 0.905 (95 % confidence interval [CI]: 0.709–1) and 0.905 (95 % CI: 0.737–1), respectively. Multi- b -value DWI-based radiomics models demonstrated strong potential for the noninvasive evaluation of tumor angiogenesis and spatial heterogeneity in HCC, suggesting that these models are promising tools for preclinical research and potential clinical translation.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Guo et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: