To develop a predictive nomogram for breast cancer lympho-vascular invasion (LVI), based on digital breast tomography (DBT) data obtained from intra- and peri-tumoral regions. One hundred ninety-two breast cancer patients were enrolled in this retrospective study from 2 institutions, in which Institution 1 served as the basis for training (n = 113) and testing (n = 49) sets, while Institution 2 served as the external validation set (n = 30). Tumor regions of interest (ROI) were manually-delineated on DBT images, in which peri-tumoral ROI was defined as 1 mm around intra-tumoral ROI. Radiomics features were extracted, and logistic regression was used to construct intra-, peri-, and intra- + peri-tumoral radiomics models. Patient clinical data was analyzed by both uni- and multi-variable logistic regression analyses to identify independent risk factors for the non-radiomics clinical imaging model, and the combination of both the most optimal radiomics and clinical imaging models comprised the comprehensive model. The best-performing model out of the 3 types (radiomics, clinical imaging, comprehensive) was identified using receiver operating characteristic (ROC) curve analysis, and used to construct the predictive nomogram. The most optimal radiomics model was the intra- + peri-tumoral model, and 3 independent risk factors for LVI, maximum tumor diameter (odds ratio [OR] = 1.486, 95% confidence interval [CI] = 1.082–2.041, P = 0.014), suspicious malignant calcification (OR = 2.898, 95% CI = 1.232 ~ 6.815, P = 0.015), and axillary lymph node (ALN) metastasis (OR = 3.615, 95% CI = 1.642–7.962, P < 0.001) were identified by the clinical imaging model. Furthermore, the comprehensive model was the most accurate in predicting LVI occurrence, with areas under the curve (AUCs) of 0.889, 0.916, and 0.862, for, respectively, the training, testing and external validation sets, compared to radiomics (0.858, 0.849, 0.844) and clinical imaging (0.743, 0.759, 0.732). The resulting nomogram, incorporating radiomics from the intra- + peri-tumoral model, as well as maximum tumor diameter, suspicious malignant calcification, and ALN metastasis, had great correspondence with actual LVI diagnoses under the calibration curve, and was of high clinical utility under decision curve analysis. The predictive nomogram, derived from both radiomics and clinical imaging features, was highly accurate in identifying future LVI occurrence in breast cancer, demonstrating its potential as an assistive tool for clinicians to devise individualized treatment regimes.

To investigate the value of applying habitat imaging (HI) radiomics on venous-phase computed tomography (CT) images from laryngeal squamous cell carcinoma (LSCC) patients, as part of a nomogram to predict Ki-67 positivity, an indicator of poorer LSCC prognoses. Clinical and CT imaging data from 128 LSCC patients, divided into training (89) and testing (39) groups, were analyzed. Conventional and HI radiomics features were extracted from enhanced venous phase images, either from the entire tumor (conventional) or 3 sub-regions (HI). Radiomics models were established, based on 5 machine learning algorithms, while clinical characteristics were analyzed by both uni- and multi-variate logistic regression analyses for their associations with Ki-67 positivity. Afterwards, a predictive nomogram was constructed by combining clinical characteristics, conventional radiomics, and HI radiomics. The only clinical characteristic strongly predictive for Ki-67-positivity is the degree of differentiation (low/medium vs. high). Additionally, HI radiomics was significantly more accurate than conventional for predicting Ki-67-positivity. The most accurate model, though, was the predictive nomogram, with areas under the curve of 0.945 (training) and 0.871 (testing), which was significantly higher than for clinical characteristics, conventional radiomics and HI radiomics models alone; it also had the highest net benefit, and thus greatest clinical utility under decision curve analysis. HI radiomics features were more accurate for predicting Ki-67-positivity in LSCC than conventional radiomics. However, the combination of those features with conventional radiomics and the degree of differentiation in a predictive nomogram yields the most accurate model for Ki-67-positivity.

Background: Lymph node metastases (LNM) in laryngeal squamous cell carcinoma (LSCC) has been associated with lower survival, but current imaging methods, such as computed tomography (CT), have limited capabilities to identify them. Both conventional radiomics, involving data analysis of high-throughput quantitative features extracted from medical images, as well as deep learning networks, improved LNM diagnostic accuracy in LSCC, but the combination of both approaches has not been fully examined. In this study, we aimed to improve LNM identification in LSCC patients by developing a predictive nomogram, combining deep learning radiomics and clinical imaging features from CT images.Methods: A retrospective analysis of 235 LSCC patients, divided into training (164) and validation (71) sets, was conducted. Radiomics features were extracted from CT images, and 7 machine learning algorithms were used to develop 7 radiomics models, which were combined with deep learning features extracted from the ResNet50 deep learning network to form deep learning radiomics (DLR) models. The optimal DLR model was combined with significant clinical imaging features from CT scans to develop the predictive nomogram for LNM in LSCC.Results: The nomogram, under receiver operating characteristic (ROC) curve analyses, yielded areas under the curve (AUC) values of, respectively, 0.934 and 0.864 for training and validation sets, significantly higher than clinical imaging features (0.832 and 0.817), conventional radiomics (0.861 and 0.818), and DLR (0.913 and 0.864), indicating that it was significantly more accurate in predicting LNM in LSCC patients. Additionally, decision curve analysis found that the nomogram had significantly higher clinical utility than the other 3 models.Conclusion: The predictive nomogram, combining clinical imaging and DLR features, is able to accurately identify LNM in LSCC patients, providing valuable information for non-invasive LN staging and personalized treatment approaches.

Background: Coronary disease that is non-obstructive at baseline may progress, while luminal stenosis alone does not capture local pericoronary tissue characteristics or lesion-specific functional impairment. Objective: To assess associations of lesion-specific pericoronary adipose tissue (PCAT) attenuation and CT-derived fractional flow reserve (CT-FFR) with imaging-defined stenosis progression and to evaluate an exploratory multivariable model. Methods: This retrospective single-center study included 201 patients undergoing serial coronary computed tomography angiography with selected target lesions classified as minimal or mild at baseline. In progressors, the lesion with the greatest follow-up progression was selected; in non-progressors, the lesion with the greatest baseline stenosis was selected. Thus, target-lesion selection was outcome dependent. Lesion-specific PCAT and CT-FFR were quantified using dedicated automated software after target-lesion identification. Five logistic models specified before reanalysis fitting were fitted in a common complete-case population ( n = 191), and Model E retained nine predictors. Internal performance was assessed with 2,000 patient-level bootstrap resamples. Results: Progression occurred in 115 of 201 patients. In Model E, less-negative PCAT attenuation (odds ratio [OR] per 10-HU increase, 1.552; 95% CI, 1.137–2.119; P = 0.006) and lower CT-FFR (OR per 0.05 decrease, 1.360; 95% CI, 1.041–1.778; P = 0.024) were associated with progression; age showed an inverse conditional association (OR per 10-year increase, 0.509; 95% CI, 0.346–0.749; P < 0.001). Model E had an apparent AUC of 0.767 (95% CI, 0.700–0.834) and an optimism-corrected AUC of 0.722 (95% bootstrap interval, 0.659–0.787). The corrected Brier score was 0.217 (0.189–0.251), calibration intercept −0.0002 (−0.354 to 0.361), and calibration slope 0.779 (0.539–1.100). Versus Model B, corrected Δ AUC was 0.053 (−0.005 to 0.088), indicating apparent improvement with uncertainty. Conclusions: Less-negative lesion-specific PCAT attenuation and lower CT-FFR were associated with imaging-defined stenosis progression. The combined model showed moderate optimism-corrected discrimination, but incremental discrimination remained uncertain. Because lesion selection in progressors incorporated follow-up information and no external validation was available, this internally derived model is exploratory and is not a clinically validated decision tool.