Development and Validation of a Radiomics Model for Differentiating Bone Islands and Osteoblastic Bone Metastases at Abdominal CT.

Background It is important to diagnose sclerotic bone lesions in order to determine treatment strategy. Purpose To evaluate the diagnostic performance of a CT radiomics-based machine learning model for differentiating bone islands and osteoblastic bone metastases. Materials and Methods In this retrospective study, patients who underwent contrast-enhanced abdominal CT and were diagnosed with a bone island or osteoblastic metastasis between 2015 to 2019 at either of two different institutions were included: institution 1 for the training set and institution 2 for the external test set. Radiomics features were extracted. The random forest (RF) model was built using 10 selected features, and subsequent 10-fold cross-validation was performed. In the test phase, the RF model was tested with an external test set. Three radiologists reviewed the CT images for the test set. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated for the models and each of the three radiologists. The AUCs of the radiomics model and radiologists were compared. Results A total of 177 patients (89 with a bone island and 88 with metastasis; mean age, 66 years ± 12 [standard deviation]; 111 men) were in the training set, and 64 (23 with a bone island and 41 with metastasis; mean age, 69 years ± 14; 59 men) were in the test set. Radiomics features (n = 1218) were extracted. The average AUC of the RF model from 10-fold cross-validation was 0.89 (sensitivity, 85% [75 of 88 patients]; specificity, 82% [73 of 89 patients]; and accuracy, 84% [148 of 177 patients]). In the test set, the AUC of the trained RF model was 0.96 (sensitivity, 80% [33 of 41 patients]; specificity, 96% [22 of 23 patients]; and accuracy, 86% [55 of 64 patients]). The AUCs for the three readers were 0.95 (95% CI: 0.90, 1.00), 0.96 (95% CI: 0.90, 1.00), and 0.88 (95% CI: 0.80, 0.96). The AUC of radiomics model was higher than that of only reader 3 (0.96 vs 0.88, respectively; P = .03). Conclusion A CT radiomics-based random forest model was proven useful for differentiating bone islands from osteoblastic metastases and showed better diagnostic performance compared with an inexperienced radiologist. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Vannier in this issue.

Prognosis Prediction in Initially Diagnosed Multiple Myeloma Patients Using Intravoxel Incoherent Motion-Diffusion Weighted Imaging and Multiecho Dixon Imaging.

BACKGROUND: Multiparametric MRI provides complementary information for the diagnosis and management of multiple myeloma (MM). PURPOSE: To evaluate the association of prognostic factors of MM and parameters derived from intravoxel-incoherent motion diffusion-weighted imaging (IVIM-DWI) and multiecho (ME) Dixon. STUDY TYPE: Retrospective. POPULATION: In all, 78 MM patients. FIELD STRENGTH/SEQUENCES: T1 -weighted turbo spin-echo sequences (TSE), IVIM-DWI, ME 3D gradient echo sequence with multistep adaptive fitting at 3T. ASSESSMENT: The region of interest (ROI) on the vertebral body was independently measured on four parametric maps (Dslow , Dfast and perfusion fraction [f], and proton-density fat-fraction [Ff] maps) by two readers. All patients were categorized into three groups based on the International Staging System (ISS). STATISTICAL TESTS: Three groups were compared using analysis of variance (ANOVA) and post-hoc tests with Bonferroni correction. Logistic regression analysis was performed to predict the advancement of disease (early vs. advanced). Principal component analysis (PCA) was used to find the deterministic parameters. RESULTS: Dslow and Ff were significantly different among ISS-1 (n = 38), ISS-2 (n = 22), and ISS-3 (n = 18) groups in both readers: 0.36, 0.41, and 0.58 × 10-3 mm2 /s for Dslow (P < 0.05), and 46%, 30%, and 15% for Ff (P < 0.05) in reader 1; 0.34, 0.41, and 0.58 × 10-3 mm2 /s for Dslow (P < 0.05), 43%, 27%, and 13.2% for Ff (P < 0.05) in reader 2, respectively. Dfast between ISS-3 and the other groups was significantly different in one reader only: 2.03, 2.29, and 2.85 × 10-3 mm2 /s (P < 0.05). There was no significant difference in f among the groups in both readers. Logistic regression by stepwise selection indicated Ff as the single most significant factor for differentiating early and advanced stages of MM with an accuracy of 76% and area under the curve (AUC) of 0.83 (P < 0.05). PCA revealed Ff, and Dslow as the deterministic parameters, with a cumulative proportion of 0.84. DATA CONCLUSION: D slow and Ff are associated with the prognostic factor of MM. Level of Evidence 3 Technical Efficacy Stage 5. J. MAGN. RESON. IMAGING 2021;53:491-501.