Three-Dimensional Multifrequency MR Elastography for Microvascular Invasion and Prognosis Assessment in Hepatocellular Carcinoma.

BACKGROUND: Pretreatment identification of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is important when selecting treatment strategies. PURPOSE: To improve models for predicting MVI and recurrence-free survival (RFS) by developing nomograms containing three-dimensional (3D) MR elastography (MRE). STUDY TYPE: Prospective. POPULATION: 188 patients with HCC, divided into a training cohort (n = 150) and a validation cohort (n = 38). In the training cohort, 106/150 patients completed a 2-year follow-up. FIELD STRENGTH/SEQUENCE: 1.5T 3D multifrequency MRE with a single-shot spin-echo echo planar imaging sequence, and 3.0T multiparametric MRI (mp-MRI), consisting of diffusion-weighted echo planar imaging, T2-weighted fast spin echo, in-phase out-of-phase T1-weighted fast spoiled gradient-recalled dual-echo and dynamic contrast-enhanced gradient echo sequences. ASSESSMENT: Multivariable analysis was used to identify the independent predictors for MVI and RFS. Nomograms were constructed for visualization. Models for predicting MVI and RFS were built using mp-MRI parameters and a combination of mp-MRI and 3D MRE predictors. STATISTICAL TESTS: Student's t-test, Mann-Whitney U test, chi-squared or Fisher's exact tests, multivariable analysis, area under the receiver operating characteristic curve (AUC), DeLong test, Kaplan-Meier analysis and log rank tests. P < 0.05 was considered significant. RESULTS: Tumor c and liver c were independent predictors of MVI and RFS, respectively. Adding tumor c significantly improved the diagnostic performance of mp-MRI (AUC increased from 0.70 to 0.87) for MVI detection. Of the 106 patients in the training cohort who completed the 2-year follow up, 34 experienced recurrence. RFS was shorter for patients with MVI-positive histology than MVI-negative histology (27.1 months vs. >40 months). The MVI predicted by the 3D MRE model yielded similar results (26.9 months vs. >40 months). The MVI and RFS nomograms of the histologic-MVI and model-predicted MVI-positive showed good predictive performance. DATA CONCLUSION: Biomechanical properties of 3D MRE were biomarkers for MVI and RFS. MVI and RFS nomograms were established. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.

DeepmRNALoc: A Novel Predictor of Eukaryotic mRNA Subcellular Localization Based on Deep Learning.

The subcellular localization of messenger RNA (mRNA) precisely controls where protein products are synthesized and where they function. However, obtaining an mRNA's subcellular localization through wet-lab experiments is time-consuming and expensive, and many existing mRNA subcellular localization prediction algorithms need to be improved. In this study, a deep neural network-based eukaryotic mRNA subcellular location prediction method, DeepmRNALoc, was proposed, utilizing a two-stage feature extraction strategy that featured bimodal information splitting and fusing for the first stage and a VGGNet-like CNN module for the second stage. The five-fold cross-validation accuracies of DeepmRNALoc in the cytoplasm, endoplasmic reticulum, extracellular region, mitochondria, and nucleus were 0.895, 0.594, 0.308, 0.944, and 0.865, respectively, demonstrating that it outperforms existing models and techniques.

Three-dimensional magnetic resonance elastography combining proton-density fat fraction precisely identifies metabolic dysfunction-associated steatohepatitis with significant fibrosis.

PURPOSE: Patients with metabolic dysfunction-associated steatohepatitis (MASH) and significant fibrosis (fibrosis stage≥2), known as Fibro-MASH, are at increased risk of liver-related outcomes and lower rates of spontaneous disease regression. The aim was to investigate three-dimensional MR elastography (3D-MRE) combining proton-density fat fraction (PDFF) as a means of identifying Fibro-MASH. METHODS: Forty-eight New Zealand rabbits were fed a high-fat/cholesterol or standard diet to obtain different disease activity and fibrosis stages. Shear stiffness (SS) and Damping Ratio (DR) were derived from 3D-MRE, whereas PDFF was from a volumetric 3D imaging sequence. Steatosis grade, metabolic dysfunction-associated steatotic liver disease activity score (MAS), and fibrosis stage were diagnosed histologically. Serum markers of fibrosis and inflammation were also measured. Correlation and comparison analysis, Receiver operating characteristic curves (ROC), Delong test, logistic regression analysis, and Net reclassification improvement (NRI) were performed. RESULTS: PDFF correlated with steatosis grade (rho = 0.853). SS increased with developed liver fibrosis (rho = 0.837). DR correlated with MAS grade (rho = 0.678). The areas under the ROC (AUROCs) of SS for fibrosis grading were 0.961 and 0.953 for ≥F2, and ≥ F3, respectively. All the biochemical parameters were considered but excluded from the logistic regression analysis to identify Fibro-MASH. FF, SS, and DR were finally included in the further analysis. The three-parameter model combining PDFF, SS, and DR showed significant improvement in NRI over the model combining SS and PDFF (AUROC 0.973 vs. 0.906, P = 0.081; NRI 0.28, P < 0.05). CONCLUSION: 3D-MRE combining PDFF may characterize the state of fat content, disease activity and fibrosis, thus precisely identify Fibro-MASH.

Three-dimensional multifrequency magnetic resonance elastography improves preoperative assessment of proliferative hepatocellular carcinoma.

BACKGROUND: To investigate the viscoelastic signatures of proliferative hepatocellular carcinoma (HCC) using three-dimensional (3D) magnetic resonance elastography (MRE). METHODS: This prospective study included 121 patients with 124 HCCs as training cohort, and validation cohort included 33 HCCs. They all underwent preoperative conventional magnetic resonance imaging (MRI) and tomoelastography based on 3D multifrequency MRE. Viscoelastic parameters of the tumor and liver were quantified as shear wave speed (c, m/s) and loss angle (φ, rad), representing stiffness and fluidity, respectively. Five MRI features were evaluated. Multivariate logistic regression analyses were used to determine predictors of proliferative HCC to construct corresponding nomograms. RESULTS: In training cohort, model 1 (Combining cirrhosis, hepatitis virus, rim APHE, peritumoral enhancement, and tumor margin) yielded an area under the curve (AUC), sensitivity, specificity, accuracy of 0.72, 58.73%,78.69%, 67.74%, respectively. When adding MRE properties (tumor c and tumor φ), established model 2, the AUC increased to 0.81 (95% CI 0.72-0.87), with sensitivity, specificity, accuracy of 71.43%, 81.97%, 75%, respectively. The C-index of nomogram of model 2 was 0.81, showing good performance for proliferative HCC. Therefore, integrating tumor c and tumor φ can significantly improve the performance of preoperative diagnosis of proliferative HCC (AUC increased from 0.72 to 0.81, p = 0.012). The same finding was observed in the validation cohort, with AUC increasing from 0.62 to 0.77 (p = 0.021). CONCLUSIONS: Proliferative HCC exhibits low stiffness and high fluidity. Adding MRE properties (tumor c and tumor φ) can improve performance of conventional MRI for preoperative diagnosis of proliferative HCC. CRITICAL RELEVANCE STATEMENT: We investigated the viscoelastic signatures of proliferative hepatocellular carcinoma (HCC) using three-dimensional (3D) magnetic resonance elastography (MRE), and find that adding MRE properties (tumor c and tumor φ) can improve performance of conventional MRI for preoperative diagnosis of proliferative HCC.

Clinic-radiomics model using liver magnetic resonance imaging helps predict chronicity of drug-induced liver injury.

BACKGROUND AND AIMS: Some drug-induced liver injury (DILI) cases may become chronic, even after drug withdrawal. Radiomics can predict liver disease progression. We established and validated a predictive model incorporating the clinical characteristics and radiomics features for predicting chronic DILI. METHODS: One hundred sixty-eight DILI patients who underwent liver gadolinium-diethylenetriamine pentaacetate-enhanced magnetic resonance imaging were recruited. The patients were clinically diagnosed using the Roussel Uclaf causality assessment method. Patients who progressed to chronicity or recovery were randomly divided into the training (70%) and validation (30%) cohorts, respectively. Hepatic T1-weighted images were segmented to extract 1672 radiomics features. Least absolute shrinkage and selection operator regression was used for feature selection, and Rad-score was constructed using support vector machines. Multivariable logistic regression analysis was performed to build a clinic-radiomics model incorporating clinical characteristics and Rad-scores. The clinic-radiomics model was evaluated for its discrimination, calibration, and clinical usefulness in the independent validation set. RESULTS: Of 1672 radiomics features, 28 were selected to develop the Rad-score. Cholestatic/mixed patterns and Rad-score were independent risk factors of chronic DILI. The clinic-radiomics model, including the Rad-score and injury patterns, distinguished chronic from recovered DILI patients in the training (area under the receiver operating characteristic curve [AUROC]: 0.89, 95% confidence interval [95% CI]: 0.87-0.92) and validation (AUROC: 0.88, 95% CI: 0.83-0.91) cohorts with good calibration and great clinical utility. CONCLUSION: The clinic-radiomics model yielded sufficient accuracy for predicting chronic DILI, providing a practical and non-invasive tool for managing DILI patients.

Accurate Identification of Antioxidant Proteins Based on a Combination of Machine Learning Techniques and Hidden Markov Model Profiles.

Antioxidant proteins (AOPs) play important roles in the management and prevention of several human diseases due to their ability to neutralize excess free radicals. However, the identification of AOPs by using wet-lab experimental techniques is often time-consuming and expensive. In this study, we proposed an accurate computational model, called AOP-HMM, to predict AOPs by extracting discriminatory evolutionary features from hidden Markov model (HMM) profiles. First, auto cross-covariance (ACC) variables were applied to transform the HMM profiles into fixed-length feature vectors. Then, we performed the analysis of variance (ANOVA) method to reduce the dimensionality of the raw feature space. Finally, a support vector machine (SVM) classifier was adopted to conduct the prediction of AOPs. To comprehensively evaluate the performance of the proposed AOP-HMM model, the 10-fold cross-validation (CV), the jackknife CV, and the independent test were carried out on two widely used benchmark datasets. The experimental results demonstrated that AOP-HMM outperformed most of the existing methods and could be used to quickly annotate AOPs and guide the experimental process.