Radiology. Imaging cancer最新文献

Purpose To develop and prospectively validate a clinical and radiologic model to predict clinically significant prostate cancer (csPCa) using biparametric MRI (bpMRI). Materials and Methods Retrospective data (acquired before March 31, 2022) from 12 medical centers were collected. Radiomic features were extracted from the whole prostate gland using segmentations generated by an automatic deep learning algorithm. A model incorporating bpMRI radiomics, age, prostate-specific antigens, the Prostate Imaging Reporting and Data System (PI-RADS), and the prostate zone lesion location was trained. A retrospective validation set and prospective data (acquired after March 31, 2022) were used to compare PI-RADS scoring (area under the receiver operating characteristic curve [AUC] and specificity at PI-RADS >3). Sensitivity analyses for sequence (T2-weighted, apparent diffusion coefficient, diffusion-weighted imaging) and scanner vendor (GE, Philips, Siemens) were performed, in addition to fairness analyses for relevant categories. Results The retrospective dataset for model development included 7157 male patients (mean age, 64.78 years; 3342 [46.7%] with csPCa), and the prospective dataset for model validation included 1629 patients (mean age, 66.19 years; 592 [36.3%] with csPCa). The multimodal model outperformed PI-RADS in the retrospective (AUC, 0.88 vs 0.80, P = .005; specificity of 71% vs 58%, P = .002) and prospective validation sets (AUC, 0.91 vs 0.85, P < .001; specificity of 77% vs 66%, P < .001), leading to 22.7% fewer biopsies compared with PI-RADS. Sensitivity analyses showed the importance of multiple sequences and vendors in achieving model generalization, as using specific sequences or vendors alone led to worse performance. Fairness analysis showed generalizability across different categories but highlighted increased sensitivity with higher PI-RADS and reduced performance in one medical center. Conclusion A multimodal model provided a temporally generalizable predictor of csPCa that outperformed PI-RADS. Keywords: Algorithm Development, Machine Learning, Model Validation, Model Training, Genital/Reproductive, Neoplasms-Primary, Oncology, Comparative Studies, Technology Assessment Supplemental material is available for this article. © RSNA, 2025.

Purpose To evaluate intravoxel incoherent motion (IVIM) biomarkers across different MRI vendors and software programs for breast cancer characterization in a two-site study. Materials and Methods This institutional review board-approved, Health Insurance Portability and Accountability Act-compliant retrospective study included 106 patients (with 18 benign and 88 malignant lesions) who underwent bilateral diffusion-weighted imaging (DWI) between February 2009 and March 2013. DWI was performed using 1.5-T (n = 6) or 3-T MRI scanners from two vendors using single-shot spin-echo echo-planar imaging or twice-refocused, bipolar gradient single-shot turbo spin-echo readout with multiple b values between 0 and 1000 sec/mm². IVIM parameters tissue diffusivity (D_t), perfusion fraction (f_p), pseudo-diffusivity (D_p), and their respective first-order radiomics were derived using two software packages (Igor; Wavemetrics, and Firevoxel; New York University). Bland-Altman analysis compared IVIM metrics from the two software programs. Histopathology was the reference standard, where logistic regressions with adjustments for site compared benign and malignant lesions. Least absolute shrinkage and selection operator (LASSO) penalized multivariable regression was performed first for metrics derived from each parameter separately, and then after incorporating metrics from all three parameters. Area under receiver operating characteristic (ROC) curve (AUC) ± standard error was used to quantify the diagnostic value. Performance was also evaluated using threefold cross-validation of the combined cohort. Results In total, 49 (mean age, 49 years ± 11 [SD]) and 57 (mean age, 48 years ± 10) female patients were enrolled from sites 1 and 2, respectively. Software 1 (Igor) and software 2 (Firevoxel) identified diagnostic biomarkers individually and in multivariable analysis. Tissue diffusivity exhibited the highest software consistency, with coefficients of variation of 4.8% and 2.8% (site 1 and site 2, respectively), followed by perfusion fraction (14.5% and 18.9%) and pseudo-diffusivity (36.9% and 19.8%). The highest performing metrics were D_t,min (AUC, 0.786 ± 0.05), f_p,max (AUC, 0.835 ± 0.04), and D_p,max (AUC, 0.804 ± 0.05) for software 1 and D_t,skew (AUC, 0.82 ± 0.05), f_p,max (AUC, 0.82 ± 0.046), and D_p,max (AUC, 0.75 ± 0.06) for software 2. Five metrics (D_t,min, D_t,skew, f_p,max, D_p,min, D_p,max) were included in the multivariable regression, achieving AUCs of 0.90 ± 0.03 and 0.90 ± 0.03 for software 1, and 0.84 ± 0.04 and 0.81 ± 0.05 for software 2, without and with cross-validation, respectively. Conclusion This study confirmed the translational potential of IVIM biomarkers for breast cancer characterization. Keyword