Hepatitis C related chronic liver cirrhosis: feasibility of texture analysis of MR images for classification of fibrosis stage and necroinflammatory activity grade

TMM: A comprehensive CAD system for hepatic fibrosis 5-grade METAVIR staging based on liver MRI

BACKGROUND

Precise staging of hepatic fibrosis with MRI is necessary as it can assist precision medicine. Computer aided diagnosis (CAD) system with distinguishing radiomics features and radiologists domain knowledge is expected to obtain 5-grade meta-analysis of histological data in viral hepatitis (METAVIR) staging.

PURPOSE

This study aims to obtain the precise staging of hepatic fibrosis based on MRI as it predicts the risk of future liver-related morbidity and the need for treatment, monitoring and surveillance. Based on METAVIR score, fibrosis can be divided into five stages. Most previous researches focus on binary classification, such as cirrhosis versus non-cirrhosis, early versus advanced fibrosis, and substantial fibrosis or not. In this paper, a comprehensive CAD system TMM is proposed to precisely class hepatic fibrosis into five stages for precision medicine instead of the common binary classification.

METHODS

We propose a novel hepatic fibrosis staging CAD system TMM which includes three modules, Two-level Image Statistical Radiomics Feature (TISRF), Monotonic Error Correcting Output Codes (MECOC) and Monotone Multiclassification with Deep Forest (MMDF). TISRF extracts radiomics features for distinguishing different hepatic fibrosis stages. MECOC is proposed to encode monotonic multiclass by making full use of the progressive severity of hepatic fibrosis and increase the fault tolerance and error correction ability. MMDF combines multiple Deep Forest network to ensure the final five-class classification, which can achieve more precise classification than the common binary classification. The performance of the proposed hepatic fibrosis CAD system is tested on the hepatic data collected from our rabbits models of fibrosis.

RESULTS

A total of 140 regions of interest (ROI) are selected from MRI T1W of liver fibrosis models in 35 rabbits with F0(n = 16), F1(n = 28), F2(n = 29), F3(n = 44) and F4(n = 23). The performance is evaluated by five-fold cross-validation. TMM can achieve the highest total accuracy of 72.14% for five fibrosis stages compared with other popular classifications. To make a comprehensive comparison, a binary classification experiment have been carried out.

CONCLUSIONS

T1WI can obtain precise staging of hepatic fibrosis with the help of comprehensive CAD including radiomics features extraction inspired by radiologists, monotonic multiclass according to the severity of hepatic fibrosis, and deep learning classification.

Hybrid model for precise hepatitis-C classification using improved random forest and SVM method

Hepatitis C Virus (HCV) is a viral infection that causes liver inflammation. Annually, approximately 3.4 million cases of HCV are reported worldwide. A diagnosis of HCV in earlier stages helps to save lives. In the HCV review, the authors used a single ML-based prediction model in the current research, which encounters several issues, i.e., poor accuracy, data imbalance, and overfitting. This research proposed a Hybrid Predictive Model (HPM) based on an improved random forest and support vector machine to overcome existing research limitations. The proposed model improves a random forest method by adding a bootstrapping approach. The existing RF method is enhanced by adding a bootstrapping process, which helps eliminate the tree's minor features iteratively to build a strong forest. It improves the performance of the HPM model. The proposed HPM model utilizes a 'Ranker method' to rank the dataset features and applies an IRF with SVM, selecting higher-ranked feature elements to build the prediction model. This research uses the online HCV dataset from UCI to measure the proposed model's performance. The dataset is highly imbalanced; to deal with this issue, we utilized the synthetic minority over-sampling technique (SMOTE). This research performs two experiments. The first experiment is based on data splitting methods, K-fold cross-validation, and training: testing-based splitting. The proposed method achieved an accuracy of 95.89% for k = 5 and 96.29% for k = 10; for the training and testing-based split, the proposed method achieved 91.24% for 80:20 and 92.39% for 70:30, which is the best compared to the existing SVM, MARS, RF, DT, and BGLM methods. In experiment 2, the analysis is performed using feature selection (with SMOTE and without SMOTE). The proposed method achieves an accuracy of 41.541% without SMOTE and 96.82% with SMOTE-based feature selection, which is better than existing ML methods. The experimental results prove the importance of feature selection to achieve higher accuracy in HCV research.

Development of an MRI-Based Radiomics-Clinical Model to Diagnose Liver Fibrosis Secondary to Pancreaticobiliary Maljunction in Children

BACKGROUND

Preoperative diagnosis of liver fibrosis in children with pancreaticobiliary maljunction (PBM) is needed to guide clinical decision-making and improve patient prognosis.

PURPOSE

To develop and validate an MR-based radiomics-clinical nomogram for identifying liver fibrosis in children with PBM.

STUDY TYPE

Retrospective.

POPULATION

A total of 136 patients with PBM from two centers (center A: 111 patients; center B: 25 patients). Cases from center A were randomly divided into training (74 patients) and internal validation (37 patients) sets. Cases from center B were assigned to the external validation set. Liver fibrosis was determined by histopathological examination.

FIELD STRENGTH/SEQUENCE

A 3.0 T (two vendors)/T1-weighted imaging and T2-weighted imaging.

ASSESSMENT

Clinical factors associated with liver fibrosis were evaluated. A total of 3562 radiomics features were extracted from segmented liver parenchyma. Maximum relevance minimum redundancy and least absolute shrinkage and selection operator were recruited to screen radiomics features. Based on the selected variables, multivariate logistic regression was used to construct the clinical model, radiomics model, and combined model. The combined model was visualized as a nomogram to show the impact of the radiomics signature and key clinical factors on the individual risk of developing liver fibrosis.

STATISTICAL TESTS

Mann-Whitney U and chi-squared tests were used to compare clinical factors. P < 0.05 was considered statistically significant in the final models.

RESULTS

Two clinical factors and four radiomics features were selected as they were associated with liver fibrosis in the training (AUC, 0.723, 0.927), internal validation (AUC, 0.718, 0.885), and external validation (AUC, 0.737, 0.865) sets. The radiomics-clinical nomogram yielded the best performance in the training (AUC, 0.977), internal validation (AUC, 0.921), and external validation (AUC, 0.878) sets, with good calibration (P > 0.05).

DATA CONCLUSION

Our radiomic-based nomogram is a noninvasive, accurate, and preoperative diagnostic tool that is able to detect liver fibrosis in PBM children.

EVIDENCE LEVEL

3.

TECHNICAL EFFICACY

Stage 2.

Predictive value of Gd-EOB-DTPA -enhanced magnetic resonance imaging for post-hepatectomy liver failure: a systematic review and meta-analysis

Background

Accurate preoperative diagnosis of post-hepatectomy liver failure (PHLF) is particularly important to improve the prognosis of patients.

Purpose

To evaluate the predictive value of Gd-EOB-DTPA-enhanced magnetic resonance imaging (MRI) for post-hepatectomy liver failure.

Material and Methods

A systematic search was performed in the PubMed, Embase, the Cochrane Library, and Web of Science databases to find relevant original articles published up to December 2021. The included studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. The bivariate random-effects model was used to assess the diagnostic authenticity. Meta-regression analyses were performed to analyze the potential heterogeneity.

Results

In total, 13 articles were included. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the area under the summary receiver operating characteristic curves were 88% (95% confidence interval [CI] = 0.80–0.94), 80% (95% CI = 0.73–0.86), 4.4 (95% CI = 3.3–5.9), 0.14 (95% CI = 0.08–0.25), 31 (95% CI = 17–57), and 0.91 (95% CI = 0.89–0.94), respectively. There was no publication bias and threshold effect in our study.

Conclusion

Gd-EOB-DTPA-enhanced MRI is a potentially useful for the prediction of PHLF after major hepatectomy.

Development of an equation to screen for solar hemorrhages from digital cushion ultrasound texture analysis in veal calves at slaughter

Claw disorders are a relevant welfare issue in the cattle industry, fast and accurate diagnoses are essential for successful treatment and prevention. The present study aimed to develop an equation to assess the presence of solar hemorrhages from real-time ultrasound images texture analysis at slaughter. Eighty-eight hind feet were collected at the slaughterhouse from 44 Holstein male veal calves. The claws were trimmed by a veterinarian hoof-trimmer, approximately 30 min after the calves' slaughter, and classified into healthy and affected by solar hemorrhages. At the same time, ultrasound images were collected for each claw. Sole soft tissues' thickness was measured, and texture analysis was performed using MaZda software. The resulting parameters from sole soft tissues' measurements and texture analysis were screened with a stepwise linear discriminant analysis using the absence or presence (0/1) of solar hemorrhages as the dependent variable. Results from the stepwise analysis identified 9 variables (among 279) as predictors, and an equation was developed and used to predict the presence or absence of solar hemorrhages on the scanned claws by binary measure: values ≤0.5 counted as 0, while those >0.5 as 1. Validation of the equation was performed by testing predicted lesions (LESpred) against the clinically evaluated lesions (LESeval) with a confusion matrix, a ROC analysis, and a precision-recall curve. Results of the present study suggest that the equation proposed has a good potential for detecting effectively hemorrhages of the sole by ultrasound imaging texture means, and could be used to monitor unsatisfactory housing and management conditions at the farm level, and for early management intervention and prevention.

The Value of First-Order Features Based on the Apparent Diffusion Coefficient Map in Evaluating the Therapeutic Effect of Low-Intensity Pulsed Ultrasound for Acute Traumatic Brain Injury With a Rat Model

Purpose

In order to evaluate the neuroprotective effect of low-intensity pulsed ultrasound (LIPUS) for acute traumatic brain injury (TBI), we studied the potential of apparent diffusion coefficient (ADC) values and ADC-derived first-order features regarding this problem.

Methods

Forty-five male Sprague Dawley rats (sham group: 15, TBI group: 15, LIPUS treated: 15) were enrolled and underwent magnetic resonance imaging. Scanning layers were acquired using a multi-shot readout segmentation of long variable echo trains (RESOLVE) to decrease distortion. The ultrasound transducer was applied to the designated region in the injured cortical areas using a conical collimator and was filled with an ultrasound coupling gel. Regions of interest were manually delineated in the center of the damaged cortex on the diffusion weighted images (b = 800 s/mm²) layer by layer for the TBI and LIPUS treated groups using the open-source software ITK-SNAP. Before analysis and modeling, the features were normalized using a z-score method, and a logistic regression model with a backward filtering method was employed to perform the modeling. The entire process was completed using the R language.

Results

During the observation time, the ADC values ipsilateral to the trauma in the TBI and LIPUS groups increased rapidly up to 24 h. After statistical analysis, the 10th percentile, 90th percentile, mean, skewness, and uniformity demonstrated a significant difference among three groups. The receiver operating characteristic curve (ROC) analysis shows that the combined LR model exhibited the highest area under the curve value (AUC: 0.96).

Conclusion

The combined LR model of first-order features based on the ADC map can acquire a higher diagnostic performance than each feature only in evaluating the neuroprotective effect of LIPUS for TBI. Models based on first-order features may have potential value in predicting the therapeutic effect of LIPUS in clinical practice in the future.

Usefulness of Noncontrast MRI-Based Radiomics Combined Clinic Biomarkers in Stratification of Liver Fibrosis

PURPOSE

To develop and validate a radiomic nomogram based on texture features from out-of-phase T1W images and clinical biomarkers in prediction of liver fibrosis.

MATERIALS AND METHODS

Patients clinically diagnosed with chronic liver fibrosis who underwent liver biopsy and noncontrast MRI were enrolled. All patients were assigned to the nonsignificant fibrosis group with fibrosis stage <2 and the significant fibrosis group with stage ≥2. Texture parameters were extracted from out-of-phase T1-weighted (T1W) images and calculated using the Artificial Intelligent Kit (AK). Boruta and LASSO regressions were used for feature selection and a multivariable logistic regression was used for construction of a combinational model integrating radiomics and clinical biomarkers. The performance of the models was assessed by using the receiver operator curve (ROC) and decision curve.

RESULTS

ROC analysis of the radiomics model that included the most discriminative features showed AUCs of the training and test groups were 0.80 and 0.78. A combinational model integrating RADscore and fibrosis 4 index was established. ROC analysis of the training and test groups showed good to excellent performance with AUC of 0.93 and 0.86. Decision curves showed the combinational model added more net benefit than radiomic and clinical models alone.

CONCLUSIONS

The study presents a combinational model that incorporates RADscore and clinical biomarkers, which is promising in classification of liver fibrosis.

Application of Ultrasound Images Texture Analysis for the Estimation of Intramuscular Fat Content in the Longissimus Thoracis Muscle of Beef Cattle after Slaughter: A Methodological Study

Simple Summary

Fat content in the muscle mass (IMF) is one of the most important characteristics influencing the aroma, tenderness, and juiciness of the meat and therefore has high importance for both commercialization purposes and consumers. However, IMF determination currently relies on visual inspection, which is a subjective and inconsistent technique. The aim of the present study is the elaboration of a procedure capable of predicting IMF% in beef carcasses using ultrasound imaging texture analysis. Ultrasound images taken on meat samples were compared to meat composition measured by chemical extraction. Determination coefficient between the two techniques was R² = 0.76, while Receiver Operating Characteristic analysis showed a sensitivity of 88% and a specificity of 90%. The results therefore suggest that the described procedure is expected to determine IMF% in muscle with good accuracy. Ultrasound imaging could be applied in routine beef grading practices. This may help to solve the issues related to subjectivity and leave to the operator only imaging acquisition. Better consistency in beef products could enhance consumers’ satisfaction and commercial standardization programs.

Abstract

Intramuscular fat (IMF) is a major trait in the evaluation of beef meat, but its determination is subjective and inconsistent and still relies on visual inspection. This research objective was a method to predict IMF% from beef meat using ultrasound (US) imaging texture analysis. US images were performed on the longissimus thoracis muscle of 27 Charolaise heifers. Cuts from the 12th to 13th ribs were scanned. The lipid content of the muscle samples was determined with the petrol ether (Randall) extraction method. A stepwise linear discriminant analysis was used to screen US texture parameters. IMF% measured by chemical extraction (IMFqa) was the dependent variable and the results of the texture analysis were the explanatory variables. The model highlighted seven parameters, as a predictive and a multiple regression equation was created. Prediction of IMF content (IMFpred) was then validated using IMFqa as ground truth. Determination coefficient between IMFqa and IMFpred was R² = 0.76, while the ROC analysis showing a sensitivity of 88% and a specificity of 90%. Bland-Altman plot upper and lower limit were +1.34 and −1.42, respectively (±1.96 SD), with a mean of −0.04. The results from the present study therefore suggest that prediction of IMF content in muscle mass by US texture analysis is possible.

Imaging-Based Staging of Hepatic Fibrosis in Patients with Hepatitis B: A Dynamic Radiomics Model Based on Gd-EOB-DTPA-Enhanced MRI

Accurate grading of liver fibrosis can effectively assess the severity of liver disease and help doctors make an appropriate diagnosis. This study aimed to perform the automatic staging of hepatic fibrosis on patients with hepatitis B, who underwent gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced magnetic resonance imaging with dynamic radiomics analysis. The proposed dynamic radiomics model combined imaging features from multi-phase dynamic contrast-enhanced (DCE) images and time-domain information. Imaging features were extracted from the deep learning-based segmented liver volume, and time-domain features were further explored to analyze the variation in features during contrast enhancement. Model construction and evaluation were based on a 132-case data set. The proposed model achieved remarkable performance in significant fibrosis (fibrosis stage S1 vs. S2–S4; accuracy (ACC) = 0.875, area under the curve (AUC) = 0.867), advanced fibrosis (S1–S2 vs. S3–S4; ACC = 0.825, AUC = 0.874), and cirrhosis (S1–S3 vs. S4; ACC = 0.850, AUC = 0.900) classifications in the test set. It was more dominant compared with the conventional single-phase or multi-phase DCE-based radiomics models, normalized liver enhancement, and some serological indicators. Time-domain features were found to play an important role in the classification models. The dynamic radiomics model can be applied for highly accurate automatic hepatic fibrosis staging.

Texture Analysis of Gray-Scale Ultrasound Images for Staging of Hepatic Fibrosis

Purpose

Materials and Methods

Results

Conclusion

Noninvasive Assessment of Liver Parenchyma Using Gray-Scale Ultrasound-Based Histogram Analysis in Patients With Chronic Hepatitis B Infection

The aims of this study were to examine the alterations of liver echo-intensity histogram parameters in chronic hepatitis B (CHB) patients and to assess the potential role of histogram parameters in the evaluation of hepatic fibrosis. A total of 52 patients with CHB who underwent liver biopsies were included in the study. The control group consisted of 30 healthy individuals. Histogram parameters were obtained from histogram analysis of gray-scale ultrasound images of both groups. The histogram parameters of the groups were compared. The association of histogram parameters with the grading and staging of histological activity index (HAI) in patients with CHB were evaluated. The patient group had statistically significant lower skewness, kurtosis, and higher variance, mean, 50th, and 90th percentile values compared with control group. When patients with CHB were divided into subgroups according to HAI stage, there was the increasing trend in skewness values and decreasing trend in kurtosis values across subgroups. The first percentile values showed negative correlation with HAI staging in patients with CHB. Ultrasound is a fast, inexpensive, and reproducible imaging method; histogram analysis of gray-scale ultrasound images may provide useful information for evaluation of hepatic fibrosis in CHB patients.

Comparison of handcrafted features and convolutional neural networks for liver MR image adequacy assessment

We propose a random forest classifier for identifying adequacy of liver MR images using handcrafted (HC) features and deep convolutional neural networks (CNNs), and analyze the relative role of these two components in relation to the training sample size. The HC features, specifically developed for this application, include Gaussian mixture models, Euler characteristic curves and texture analysis. Using HC features outperforms the CNN for smaller sample sizes and with increased interpretability. On the other hand, with enough training data, the combined classifier outperforms the models trained with HC features or CNN features alone. These results illustrate the added value of HC features with respect to CNNs, especially when insufficient data is available, as is often found in clinical studies.

Development and validation of radiomics model built by incorporating machine learning for identifying liver fibrosis and early-stage cirrhosis

BACKGROUND

Liver fibrosis (LF) continues to develop and eventually progresses to cirrhosis. However, LF and early-stage cirrhosis (ESC) can be reversed in some cases, while advanced cirrhosis is almost impossible to cure. Advances in quantitative imaging techniques have made it possible to replace the gold standard biopsy method with non-invasive imaging, such as radiomics. Therefore, the purpose of this study is to develop a radiomics model to identify LF and ESC.

METHODS

Patients with LF (n = 108) and ESC (n = 116) were enrolled in this study. As a control, patients with healthy livers were involved in the study (n = 145). Diffusion-weighted imaging (DWI) data sets with three b-values (0, 400, and 800 s/mm) of enrolled cases were collected in this study. Then, radiomics features were extracted from manually delineated volumes of interest. Two modeling strategies were performed after univariate analysis and feature selection. Finally, an optimal model was determined by the receiver operating characteristic area under the curve (AUC).

RESULTS

The optimal models were built in plan 1. For model 1 in plan 1, the AUCs of the training and validation cohorts were 0.973 (95% confidence interval [CI] 0.946-1.000) and 0.948 (95% CI 0.903-0.993), respectively. For model 2 in plan 1, the AUCs of the training and validation cohorts were 0.944, 95% CI 0.905 to 0.983, and 0.968, 95% CI 0.940 to 0.996, respectively.

CONCLUSIONS

Radiomics analysis of DWI images allows for accurate identification of LF and ESC, and the non-invasive biomarkers extracted from the functional DWI images can serve as a better alternative to biopsy.

Radiomics model based on preoperative gadoxetic acid-enhanced MRI for predicting liver failure

BACKGROUND

Postoperative liver failure is the most severe complication in cirrhotic patients with hepatocellular carcinoma (HCC) after major hepatectomy. Current available clinical indexes predicting postoperative residual liver function are not sufficiently accurate.

AIM

To determine a radiomics model based on preoperative gadoxetic acid-enhanced magnetic resonance imaging for predicting liver failure in cirrhotic patients with HCC after major hepatectomy.

METHODS

For this retrospective study, a radiomics-based model was developed based on preoperative hepatobiliary phase gadoxetic acid-enhanced magnetic resonance images in 101 patients with HCC between June 2012 and June 2018. Sixty-one radiomic features were extracted from hepatobiliary phase images and selected by the least absolute shrinkage and selection operator method to construct a radiomics signature. A clinical prediction model, and radiomics-based model incorporating significant clinical indexes and radiomics signature were built using multivariable logistic regression analysis. The integrated radiomics-based model was presented as a radiomics nomogram. The performances of clinical prediction model, radiomics signature, and radiomics-based model for predicting post-operative liver failure were determined using receiver operating characteristics curve, calibration curve, and decision curve analyses.

RESULTS

Five radiomics features from hepatobiliary phase images were selected to construct the radiomics signature. The clinical prediction model, radiomics signature, and radiomics-based model incorporating indocyanine green clearance rate at 15 min and radiomics signature showed favorable performance for predicting postoperative liver failure (area under the curve: 0.809-0.894). The radiomics-based model achieved the highest performance for predicting liver failure (area under the curve: 0.894; 95%CI: 0.823-0.964). The integrated discrimination improvement analysis showed a significant improvement in the accuracy of liver failure prediction when radiomics signature was added to the clinical prediction model (integrated discrimination improvement = 0.117, P = 0.002). The calibration curve and an insignificant Hosmer-Lemeshow test statistic (P = 0.841) demonstrated good calibration of the radiomics-based model. The decision curve analysis showed that patients would benefit more from a radiomics-based prediction model than from a clinical prediction model and radiomics signature alone.

CONCLUSION

A radiomics-based model of preoperative gadoxetic acid–enhanced MRI can be used to predict liver failure in cirrhotic patients with HCC after major hepatectomy.

Whole-Volume ADC Histogram Analysis in Parotid Glands to Identify Patients with Sjögren's Syndrome

At present, no gold standard for diagnosing Sjögren’s syndrome (SS) is available in clinical practice. The 2002 American–European Consensus Group classification criteria are used to diagnose SS. Clinically, it is challenging to distinguish patients with SS from suspected patients undergoing different therapies. A total of 52 patients with SS and 24 patients suspected of having the disease prospectively underwent 3.0-T magnetic resonance (MR) scanning, including diffusion-weighted imaging (b = 0 and 1000 s/mm²). The whole-volume apparent diffusion coefficient (ADC) histogram analysis generated ADC_mean, skewness, kurtosis, and entropy values from bilateral parotid glands. Continuous variables were compared using an independent two-sample t test, and categorical variable compared using the Fisher’s test between the two groups. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of the indexes. Fisher’s tests demonstrated that some clinical indexes and MR morphology grades differed significantly between patients with SS and patients suspected of having the disease (all P ≤ 0.001). The parotid entropy value of patients with SS was significantly higher than that of patients suspected of having the disease (P < 0.001). Among MR parameters, entropy combined with kurtosis performed the best in differentiating patients with SS from those suspected of having SS (area under the ROC curve = 0.955). A whole-volume ADC histogram analysis might provide a series of parameters that reflect tissue characteristics.

Diagnostic value of MR-based texture analysis for the assessment of hepatic fibrosis in patients with nonalcoholic fatty liver disease (NAFLD)

PURPOSE

To investigate the performance of MR-based texture analysis (TA) for the assessment of hepatic fibrosis in patients with nonalcoholic fatty liver disease (NAFLD).

METHODS

Fifty-four adult patients (33 females, 21 males, mean age 49.8 ± 13.5 years) with biopsy-proven NAFLD were enrolled and underwent MR imaging on a 1.5 T system. TA parameters were extracted on axial noncontrast 3D-GRE T1W images (slice thickness = 4.6 mm) using a commercially available research software (TexRAD). Receiver operating curves (ROC), areas under the ROC (AUROC) and 95% confidence intervals (CI) were calculated to assess the accuracy of each TA parameter for the diagnosis of significant (F ≥ 2) and advanced fibrosis (F ≥ 3). The correlation between TA and histopathological features of nonalcoholic steatohepatitis (NASH) was tested calculating the Spearman's rank correlation coefficient (ρ).

RESULTS

Thirty-seven (68%) subjects had significant fibrosis and 20 (37%) had advanced fibrosis. The TA parameters with the best performance were standard deviation (SD) and entropy, respectively, with AUROC 0.755 (95% CI 0.619-0.862, p ≤ 0.0002) and 0.769 (95% CI 0.634-0.873, p < 0.0001) for significant fibrosis and AUROC 0.746 (95% CI 0.609-0.854, p ≤ 0.0004) and 0.754 (95% CI 0.618-0.861, p ≤ 0.0002) for advanced fibrosis. SD and entropy demonstrated a moderate correlation with the degree of fibrosis (ρ = 0.457 and 0.480; p < 0.01). No significant correlation was found between TA parameters and other histopathological features of NASH.

CONCLUSIONS

Entropy and SD extracted on T1-weighted MR images have fair accuracy for the diagnosis of significant and advanced hepatic fibrosis in patients with NAFLD.

Radiomics Analysis of Gadoxetic Acid-enhanced MRI for Staging Liver Fibrosis

Purpose To develop and validate a radiomics-based model for staging liver fibrosis by using gadoxetic acid-enhanced hepatobiliary phase MRI. Materials and Methods In this retrospective study, 436 patients (mean age, 51 years; age range, 18-86 years; 319 men [mean age, 51 years; age range, 18-86 years]; 117 women [mean age, 50 years; age range, 18-79 years]) with pathologic analysis-proven liver fibrosis who underwent gadoxetic acid-enhanced MRI from June 2015 to December 2016 were randomized in a three-to-one ratio into development (n = 329) and test (n = 107) cohorts, respectively. In the development cohort, a model was developed to calculate radiomics fibrosis index (RFI) by using logistic regression with elastic net regularization to differentiate stage F3-F4 from stage F0-F2. Optimal RFI cutoffs to diagnose clinically significant fibrosis (stage F2-F4), advanced fibrosis (stage F3-F4), and cirrhosis (stage F4) were determined by receiver operating characteristic curve analysis. In the test cohort, the diagnostic performance of RFI was compared with that of normalized liver enhancement, aspartate transaminase-to-platelet ratio index (APRI), and fibrosis-4 index by using the Obuchowski index. Results In the test cohort, RFI (Obuchowski index, 0.86) significantly outperformed normalized liver enhancement (Obuchowski index, 0.77; P < .03), APRI (Obuchowski index, 0.60; P < .001), and fibrosis-4 index (Obuchowski index, 0.62; P < .001) for staging liver fibrosis. By using the cutoffs, RFI had sensitivities and specificities as follows: 81% (95% confidence interval: 71%, 89%) and 78% (95% confidence interval: 63%, 89%) for diagnosing stage F2-F4, respectively; 79% (95% confidence interval: 67%, 88%) and 82% (95% confidence interval: 69%, 91%), respectively, for diagnosing stage F3-F4; and 92% (95% confidence interval: 79%, 98%) and 75% (95% confidence interval: 62%, 83%), respectively, for diagnosing stage F4. Conclusion Radiomics analysis of gadoxetic acid-enhanced hepatobiliary phase images allows for accurate diagnosis of liver fibrosis. © RSNA, 2018 Online supplemental material is available for this article.

Whole-volume ADC Histogram and Texture Analyses of Parotid Glands as an Image Biomarker in Evaluating Disease Activity of Primary Sjögren's Syndrome

Diffusion weighted imaging (DWI) has proven to be sensitive for detecting early injury to the parotid gland in pSS (primary Sjögren’s syndrome). Here, we explored the application of ADC histogram and texture analyses for evaluating the disease activity of pSS. A total of 55 patients with pSS who met the classification criteria of the 2002 AECG criteria prospectively underwent 3.0-T magnetic resonance imaging (MRI) including DWI (b = 0 and 1000 s/mm²). According to the ESSDAI score, 35 patients were categorized into the low-activity group (ESSDAI < 5) and 20 into the moderate-high-activity group (ESSDAI ≥ 5). Via analysis of the whole-volume ADC histogram, the ADC_mean, skewness, kurtosis, and entropy values of the bilateral parotid glands were determined. Multivariate analysis was used to identify independent risk factors for predicting disease activity. The diagnostic performance of the indexes was evaluated via receiver operating characteristic (ROC) analysis. ROC analysis showed that the anti-SSB, lip biopsy, MRI morphology, ADC, ADC_mean, and entropy values were able to categorize the disease into two groups, particularly the entropy values. The multivariate model, which included anti-SSB, MRI morphology and entropy, had an area under the ROC curve of 0.923 (P < 0.001). The parotid entropy value distinguished disease activity in patients with pSS, especially combined with anti-SSB and MRI morphology.

Adrenal incidentaloma: machine learning-based quantitative texture analysis of unenhanced CT can effectively differentiate sPHEO from lipid-poor adrenal adenoma

Objective: To evaluate the feasibility and accuracy of machine learning based texture analysis of unenhanced CT images in differentiating subclinical pheochromocytoma (sPHEO) from lipid-poor adenoma (LPA) in adrenal incidentaloma (AI).

Methods: Seventy-nine patients with 80 LPA and 29 patients with 30 sPHEO were included in the study. Texture parameters were derived using imaging software (MaZda). Thirty texture features were selected and LPA was performed for the features selected. The number of positive features was used to predict results. Logistic multiple regression analysis was performed on the 30 texture features, and a predictive equation was created based on the coefficients obtained.

Results: LPA yielded a misclassification rate of 19.39% in differentiating sPHEO from LPA. Our predictive model had an accuracy rate of 94.4% (102/108), with a sensitivity of 86.2% (25/29) and a specificity of 97.5% (77/79) for differentiation. When the number of positive features was greater than 8, the accuracy of prediction was 85.2% (92/108), with a sensitivity of 96.6% (28/29) and a specificity of 81% (64/79).

Conclusions: Machine learning-based quantitative texture analysis of unenhanced CT may be a reliable quantitative method in differentiating sPHEO from LPA when AI is present.

Influence of CT acquisition and reconstruction parameters on radiomic feature reproducibility

High-dimensional imaging features extracted from diagnostic imaging, called radiomics, are increasingly reported for diagnosis, prognosis, and response to therapy. Establishing the sensitivity of radiomic features to variation in scan protocols is necessary because acquisition and reconstruction parameters can vary widely across and within institutions. Our objective was to assess the reproducibility of radiomic features derived from computed tomography (CT) images by varying tube current (mA), noise index, and reconstruction [adaptive statistical iterative reconstruction (ASiR)], parameters increasingly varied by institutions seeking to reduce radiation dose in their patients. We extracted radiomic features from CT images of a uniform water phantom, anthropomorphic phantom, and a human scan. Scans were acquired from the phantoms with six tube currents (50, 100, 200, 300, 400, and 500 mA) and five noise index levels (12, 14, 16, 18, and 20), respectively. Scans of the phantoms and patient were reconstructed from 0% ASiR (i.e., filtered back projection) to 100% ASiR in increments of 10%. Two hundred and forty-eight well-known radiomic features were extracted from all scans. The concordance correlation coefficient was used to assess agreement of features. Our analysis suggests that image acquisition parameters (tube current, noise index) as well as the reconstruction technique strongly influence radiomic feature reproducibility and demonstrate a subset of reproducible features potentially usable in clinical practice.

Imaging prediction of nonalcoholic steatohepatitis using computed tomography texture analysis

OBJECTIVES

To determine if texture analysis of non-contrast-enhanced CT (NECT) images is able to predict nonalcoholic steatohepatitis (NASH).

METHODS

NECT images from 88 patients who underwent a liver biopsy for the diagnosis of suspected NASH were assessed and texture feature parameters were obtained without and with filtration. The patient population was divided into a predictive learning dataset and a validation dataset, and further divided into groups according to the prediction of liver fibrosis as assessed by hyaluronic acid levels. The reference standard was the histological result of a liver biopsy. A predictive model for NASH was developed using parameters derived from the learning dataset that demonstrated areas under the receiver operating characteristic curve (AUC) of >0.65. The resulting model was then applied to the validation dataset.

RESULTS

In patients without suspected fibrosis, the texture parameter mean without filter and skewness with a 2-mm filter were selected for the NASH prediction model. The AUC of the predictive model for the validation dataset was 0.94 and the accuracy was 94%. In patients with suspicion of fibrosis, the mean without filtration and kurtosis with a 4-mm filter were selected for the NASH prediction model. The AUC for the validation dataset was 0.60 and the accuracy was 42%.

CONCLUSIONS

In patients without suspicion of fibrosis, NECT texture analysis effectively predicted NASH.

KEY POINTS

• In patients without suspicion of fibrosis, NECT texture analysis effectively predicted NASH. • The mean without filtration and skewness with a 2-mm filter were modest predictors of NASH in patients without suspicion of liver fibrosis. • Hepatic fibrosis masks the characteristic texture features of NASH.

Liver Fibrosis: Deep Convolutional Neural Network for Staging by Using Gadoxetic Acid-enhanced Hepatobiliary Phase MR Images

Purpose To investigate the performance of a deep convolutional neural network (DCNN) model in the staging of liver fibrosis using gadoxetic acid-enhanced hepatobiliary phase magnetic resonance (MR) imaging. Materials and Methods This retrospective study included patients for whom input data (hepatobiliary phase MR images, static magnetic field of the imaging unit, and hepatitis B and C virus testing results available, either positive or negative) and reference standard data (liver fibrosis stage evaluated from biopsy or surgical specimens obtained within 6 months of the MR examinations) were available were assigned to the training (534 patients) or the test (100 patients) group. For the training group (54, 53, 81, 113, and 233 patients with fibrosis stages F0, F1, F2, F3, and F4, respectively; mean patient age, 67.4 ± 9.7 years; 388 men and 146 women), MR images with three different section levels were augmented 90-fold (rotated, parallel-shifted, brightness-changed and contrast-changed images were generated; a total of 144 180 images). Supervised training was performed by using the DCNN model to minimize the difference between the output data (fibrosis score obtained through deep learning [F_DL score]) and liver fibrosis stage. The performance of the DCNN model was evaluated in the test group (10, 10, 15, 20, and 45 patients with fibrosis stages F0, F1, F2, F3, and F4, respectively; mean patient age, 66.8 years ± 10.7; 71 male patients and 29 female patients) with receiver operating characteristic (ROC) analyses. Results The F_DL score was correlated significantly with fibrosis stage (Spearman rank correlation coefficient: 0.63; P < .001). Fibrosis stages F4, F3, and F2 were diagnosed with areas under the ROC curve of 0.84, 0.84, and 0.85, respectively. Conclusion The DCNN model exhibited a high diagnostic performance in the staging of liver fibrosis. ^© RSNA, 2017 Online supplemental material is available for this article.

Evaluation of hepatic fibrosis: a review from the society of abdominal radiology disease focus panel

Hepatic fibrosis is potentially reversible; however early diagnosis is necessary for treatment in order to halt progression to cirrhosis and development of complications including portal hypertension and hepatocellular carcinoma. Morphologic signs of cirrhosis on ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) alone are unreliable and are seen with more advanced disease. Newer imaging techniques to diagnose liver fibrosis are reliable and accurate, and include magnetic resonance elastography and US elastography (one-dimensional transient elastography and point shear wave elastography or acoustic radiation force impulse imaging). Research is ongoing with multiple other techniques for the noninvasive diagnosis of hepatic fibrosis, including MRI with diffusion-weighted imaging, hepatobiliary contrast enhancement, and perfusion; CT using perfusion, fractional extracellular space techniques, and dual-energy, contrast-enhanced US, texture analysis in multiple modalities, quantitative mapping, and direct molecular imaging probes. Efforts to advance the noninvasive imaging assessment of hepatic fibrosis will facilitate earlier diagnosis and improve patient monitoring with the goal of preventing the progression to cirrhosis and its complications.

Computed Tomography Image Texture: A Noninvasive Prognostic Marker of Hepatic Recurrence After Hepatectomy for Metastatic Colorectal Cancer

BACKGROUND

Recurrence after resection of colorectal liver metastases (CRLMs) occurs in up to 75% of patients. Preoperative prediction of hepatic recurrence may inform therapeutic strategies at the time of initial resection. Texture analysis (TA) is an established technique that quantifies pixel intensity variations (heterogeneity) on cross-sectional imaging. We hypothesized that tumoral and parenchymal changes that are predictive of overall survival (OS) and recurrence in the future liver remnant (FLR) can be detected using TA on preoperative computed tomography (CT) images.

METHODS

Patients who underwent resection for CRLM between 2003 and 2007 with appropriate preoperative CT scans were included (n = 198) in this retrospective study. Texture features extracted from the tumor and FLR, and clinicopathologic variables, were incorporated into a multivariable survival model.

RESULTS

Quantitative imaging features of the FLR were an independent predictor of both OS and hepatic disease-free survival (HDFS). Tumor texture showed significant association with OS. TA of the FLR allowed patient stratification into two groups, with significantly different risks of hepatic recurrence (hazard ratio 2.09, 95% confidence interval 1.33-3.28; p = 0.001). Patients with homogeneous parenchyma had approximately twice the risk of hepatic recurrence (41 vs. 20%).

CONCLUSION

TA of the tumor and FLR are independently associated with OS, and TA of the FLR is independently associated with HDFS. Patients with homogeneous parenchyma had a significantly higher risk of hepatic recurrence. Preoperative TA of the liver represents a potential biomarker to identify patients at risk of liver recurrence after resection for CRLM.

Non-invasive Markers of Liver Fibrosis: Adjuncts or Alternatives to Liver Biopsy?

Liver fibrosis reflects sustained liver injury often from multiple, simultaneous factors. Whilst the presence of mild fibrosis on biopsy can be a reassuring finding, the identification of advanced fibrosis is critical to the management of patients with chronic liver disease. This necessity has lead to a reliance on liver biopsy which itself is an imperfect test and poorly accepted by patients. The development of robust tools to non-invasively assess liver fibrosis has dramatically enhanced clinical decision making in patients with chronic liver disease, allowing a rapid and informed judgment of disease stage and prognosis. Should a liver biopsy be required, the appropriateness is clearer and the diagnostic yield is greater with the use of these adjuncts. While a number of non-invasive liver fibrosis markers are now used in routine practice, a steady stream of innovative approaches exists. With improvement in the reliability, reproducibility and feasibility of these markers, their potential role in disease management is increasing. Moreover, their adoption into clinical trials as outcome measures reflects their validity and dynamic nature. This review will summarize and appraise the current and novel non-invasive markers of liver fibrosis, both blood and imaging based, and look at their prospective application in everyday clinical care.