Radiogenomics nomogram based on MRI and microRNAs to predict microvascular invasion of hepatocellular carcinoma.

Purpose: This study aimed to develop and validate a radiogenomics nomogram for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC) on the basis of MRI and microRNAs (miRNAs). Materials and methods: This cohort study included 168 patients (training cohort: n = 116; validation cohort: n = 52) with pathologically confirmed HCC, who underwent preoperative MRI and plasma miRNA examination. Univariate and multivariate logistic regressions were used to identify independent risk factors associated with MVI. These risk factors were used to produce a nomogram. The performance of the nomogram was evaluated by receiver operating characteristic curve (ROC) analysis, sensitivity, specificity, accuracy, and F1-score. Decision curve analysis was performed to determine whether the nomogram was clinically useful. Results: The independent risk factors for MVI were maximum tumor length, rad-score, and miRNA-21 (all P < 0.001). The sensitivity, specificity, accuracy, and F1-score of the nomogram in the validation cohort were 0.970, 0.722, 0.884, and 0.916, respectively. The AUC of the nomogram was 0.900 (95% CI: 0.808-0.992) in the validation cohort, higher than that of any other single factor model (maximum tumor length, rad-score, and miRNA-21). Conclusion: The radiogenomics nomogram shows satisfactory predictive performance in predicting MVI in HCC and provides a feasible and practical reference for tumor treatment decisions.

Interpretable machine learning based on CT-derived extracellular volume fraction to predict pathological grading of hepatocellular carcinoma.

PURPOSE: To develop a non-invasive auxiliary assessment method based on CT-derived extracellular volume (ECV) to predict the pathological grading (PG) of hepatocellular carcinoma (HCC). METHODS: The study retrospectively analyzed 238 patients who underwent HCC resection surgery between January 2013 and April 2023. Six machine learning algorithms were employed to construct predictive models for HCC PG: logistic regression, extreme gradient boosting, Light Gradient Boosting Machine (LightGBM), random forest, adaptive boosting, and Gaussian naive Bayes. Model performance was evaluated using receiver operating characteristic curve analysis, including area under the curve (AUC), sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and F1 score. Calibration plots were used for visual evaluation of model calibration. Clinical decision curve analysis was performed to assess potential clinical utility by calculating net benefit. RESULTS: 166 patients from Hospital A were allocated to the training set, while 72 patients from Hospital B (constituting 30.25% of the total sample) were assigned to the test set. The model achieved an AUC of 1.000 (95%CI: 1.000-1.000) in the training set and 0.927 (95%CI: 0.837-0.999) in the validation set, respectively. Ultimately, the model achieved an AUC of 0.909 (95%CI: 0.837-0.980) in the test set, with an accuracy of 0.778, sensitivity of 0.906, specificity of 0.789, negative predictive value of 0.556, and F1 score of 0.908. CONCLUSION: This study successfully developed and validated a non-invasive auxiliary assessment method based on CT-derived ECV to predict the HCC PG, providing important supplementary information for clinical decision-making.

MMP-3-1612 polymorphism - a risk factor for deep venous thrombosis formation.

BACKGROUND: We investigated the association of the 5A/6A polymorphism in the promoter region at -1612 of the matrix metalloproteinase-3 gene (MMP-3-1612) and deep venous thrombosis (DVT). PATIENTS, MATERIALS AND METHODS: The distribution of the MMP-3 (-1612 5A/6A) polymorphism in the case and control groups was detected by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). Serum MMP-3 level of two groups was detected using enzyme-linked immunosorbent assay (ELISA). HepG2 cells containing MMP-3-1612 recombinant plasmid were cultured in vitro and the MMP-3 level was defined by luminescence intensity of luciferase. A DVT rat model was built. Serum MMP-3 level in the rats' wounded vein at different time points was detected by ELISA and recorded for investigation of the association between MMP-3 and DVT. Statistical data analysis was conducted with SPSS18.0. RESULTS: On the basis of the observation of MMP-3-1612 genotype frequency and allele frequency in the case and control groups, we identified significantly higher MMP-3-1612 5A allele frequency and higher serum MMP-3 level in the case group than in the control group (both P < 0.05). According to in vitro luciferase measurements, the 5A allele had higher transcriptional activity than the 6A allele. As observed in the rat model, serum MMP-3 level increased with time passing and thrombosis formation after modelling. CONCLUSIONS: The MMP-3-1612 5A/6A polymorphism may effect serum MMP-3 level and over-expression of serum MMP-3 level may be a risk factor for DVT formation.

MDCT Evaluation of Costal Bone Lesions: Comparison of Axial, Multiplanar, and 3D Volume-Rendered Images: A Retrospective Study.

The aim of this retrospective study was to compare accuracies of axial, multiplanar, and volume-rendered 3-dimensional (3D) images in the diagnosis of costal bone lesions.Forty-one patients, aged from 10 to 72-years old, with costal bone lesions underwent multidetector CT (MDCT). Axial, multiplanar, and 3D-volume-rendered images were reviewed by 3 reviewers for the property of the lesions (fracture, tumor, and tumor-like lesions or inflammation). In case of fracture, the diagnosis was demonstrated with the location of the fracture and the amounts of the costal bone involved. In case of a tumor or tumor-like lesions, the diagnosis was demonstrated pathological property. Final diagnosis was determined by biopsy or surgery. Diagnostic accuracy and interreviewers agreement were evaluated.For the diagnosis of fractures, average accuracy was 77%, 100%, and 100% for axial, multiplanar, and 3D-volume-rendered images, respectively. For the diagnosis of tumor and tumor-like lesions, average accuracy was 90% for axial, 96% for multiplanar, and 99% for 3D-volume-rendered images. For the diagnosis of inflammation lesions, average accuracy was 100% for all the 3 image formats. Interobserver agreement independence of imaging formats was high.Multiplanar and 3D-volume-rendered images were superior to axial images in diagnosis of fracture, tumor, and tumor-like lesions; however, for the evaluation of inflammation lesions, there were no difference by 3 image formats.