US features of liver surface nodularity as a predictor of severe fibrosis in chronic hepatitis C

Noninvasive Assessment of Liver Fibrosis and Inflammation in Chronic Hepatitis B: A Dual-task Convolutional Neural Network (DtCNN) Model Based on Ultrasound Shear Wave Elastography

BACKGROUND AND AIMS

Liver stiffness (LS) measured by shear wave elastography (SWE) is often influenced by hepatic inflammation. The aim was to develop a dual-task convolutional neural network (DtCNN) model for the simultaneous staging of liver fibrosis and inflammation activity using 2D-SWE.

METHODS

A total of 532 patients with chronic hepatitis B (CHB) were included to develop and validate the DtCNN model. An additional 180 consecutive patients between December 2019 and April 2021 were prospectively included for further validation. All patients underwent 2D-SWE examination and serum biomarker assessment. A DtCNN model containing two pathways for the staging of fibrosis and inflammation was used to improve the classification of significant fibrosis (≥F2), advanced fibrosis (≥F3) as well as cirrhosis (F4).

RESULTS

Both fibrosis and inflammation affected LS measurements by 2D-SWE. The proposed DtCNN performed the best among all the classification models for fibrosis stage [significant fibrosis AUC=0.89 (95% CI: 0.87-0.92), advanced fibrosis AUC=0.87 (95% CI: 0.84-0.90), liver cirrhosis AUC=0.85 (95% CI: 0.81-0.89)]. The DtCNN-based prediction of inflammation activity achieved AUCs of 0.82 (95% CI: 0.78-0.86) for grade ≥A1, 0.88 (95% CI: 0.85-0.90) grade ≥A2 and 0.78 (95% CI: 0.75-0.81) for grade ≥A3, which were significantly higher than the AUCs of the single-task groups. Similar findings were observed in the prospective study.

CONCLUSIONS

The proposed DtCNN improved diagnostic performance compared with existing fibrosis staging models by including inflammation in the model, which supports its potential clinical application.

Photoacoustic Imaging for Assessing Tissue Oxygenation Changes in Rat Hepatic Fibrosis

Chronic liver inflammation progressively evokes fibrosis and cirrhosis resulting in compromised liver function, and often leading to cancer. Early diagnosis and staging of fibrosis is crucial because the five-year survival rate of early-stage liver cancer is high. This study investigates the progression of hepatic fibrosis induced in rats following ingestion of diethylnitrosamine (DEN). Changes in oxygen saturation and hemoglobin concentration resulting from chronic inflammation were assayed longitudinally during DEN ingestion by photoacoustic imaging (PAI). Accompanying liver tissue changes were monitored simultaneously by B-mode sonographic imaging. Oxygen saturation and hemoglobin levels in the liver increased over 5 weeks and peaked at 10 weeks before decreasing at 13 weeks of DEN ingestion. The oxygenation changes were accompanied by an increase in hepatic echogenicity and coarseness in the ultrasound image. Histology at 13 weeks confirmed the development of severe fibrosis and cirrhosis. The observed increase in PA signal representing enhanced blood oxygenation is likely an inflammatory physiological response to the dietary DEN insult that increases blood flow by the development of neovasculature to supply oxygen to a fibrotic liver during the progression of hepatic fibrosis. Assessment of oxygenation by PAI may play an important role in the future assessment of hepatic fibrosis.

ACR Appropriateness Criteria® Jaundice

Jaundice is the end result of myriad causes, which makes the role of imaging in this setting particularly challenging. In the United States, the most common causes of all types of jaundice fall into four categories including hepatitis, alcoholic liver disease, blockage of the common bile duct by a gallstone or tumor, and toxic reaction to a drug or medicinal herb. Clinically, differentiating between the various potential etiologies of jaundice requires a detailed history, targeted physical examination, and pertinent laboratory studies, the results of which allow the physician to categorize the type of jaundice into mechanical or nonmechanical causes. Imaging modalities used to evaluate the jaundiced patient (all etiologies) include abdominal ultrasound (US), CT, MR cholangiopancreatography, endoscopic retrograde cholangiopancreatography and endoscopic US. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Tomographic Ultrasound and LED-Based Photoacoustic System for Preclinical Imaging

Small animals are widely used as disease models in medical research. Noninvasive imaging modalities with functional capability play an important role in studying the disease state and treatment progress. Photoacoustics, being a noninvasive and functional modality, has the potential for small-animal imaging. However, the conventional photoacoustic tomographic systems use pulsed lasers, making it expensive, bulky, and require long acquisition time. In this work, we propose the use of photoacoustic and ultrasound tomographic imaging with LEDs as the light source and acoustic detection using a linear transducer array. We have demonstrated full-view tomographic imaging of a euthanized mouse and a potential application in liver fibrosis research.

Deep learning with ultrasonography: automated classification of liver fibrosis using a deep convolutional neural network

OBJECTIVES

The aim of this study was to develop a deep convolutional neural network (DCNN) for the prediction of the METAVIR score using B-mode ultrasonography images.

METHODS

Datasets from two tertiary academic referral centers were used. A total of 13,608 ultrasonography images from 3446 patients who underwent surgical resection, biopsy, or transient elastography were used for training a DCNN for the prediction of the METAVIR score. Pathological specimens or estimated METAVIR scores derived from transient elastography were used as a reference standard. A four-class model (F0 vs. F1 vs. F23 vs. F4) was developed. Diagnostic performance of the algorithm was validated on a separate internal test set of 266 patients with 300 images and external test set of 572 patients with 1232 images. Performance in classification of cirrhosis was compared between the DCNN and five radiologists.

RESULTS

The accuracy of the four-class model was 83.5% and 76.4% on the internal and external test set, respectively. The area under the receiver operating characteristic curve (AUC) for classification of cirrhosis (F4) was 0.901 (95% confidence interval [CI], 0.865-0.937) on the internal test set and 0.857 (95% CI, 0.825-0.889) on the external test set, respectively. The AUC of the DCNN for classification of cirrhosis (0.857) was significantly higher than that of all five radiologists (AUC range, 0.656-0.816; p value < 0.05) using the external test set.

CONCLUSIONS

The DCNN showed high accuracy for determining METAVIR score using ultrasonography images and achieved better performance than that of radiologists in the diagnosis of cirrhosis.

KEY POINTS

• DCNN accurately classified the ultrasonography images according to the METAVIR score. • The AUROC of this algorithm for cirrhosis assessment was significantly higher than that of radiologists. • DCNN using US images may offer an alternative tool for monitoring liver fibrosis.

Ultrasound is highly specific in diagnosing compensated cirrhosis in chronic hepatitis C patients in real world clinical practice

Ultrasound is routinely used during the evaluation of liver cirrhosis. Inter-observer variability is considered a major drawback. This retrospective study investigated the accuracy of ultrasound in diagnosing compensated cirrhosis (i.e., modified Knodell F3, F4) in chronic hepatitis C (CHC) patients in real world clinical practice. Consecutive treatment-naive CHC patients who underwent liver biopsy (LB) prior to interferon therapy from 1997 to 2010 were enrolled. Ultrasound was performed by 30 hepatologists prior to LB. Ultrasound-identified cirrhosis was defined as small liver size, nodular liver surface and coarse liver parenchyma. LB was used as a reference, and the diagnostic accuracy of ultrasound was assessed and compared. Fibrosis was scored according to the modified Knodell classification. A cohort comprising 1738 patients, including 922 men and 816 women with a mean age of 52.5 years, was analyzed in the present study. The distribution of the patients' modified Knodell scores was F0 = 336, F1 = 489, F2 = 165, F3 = 315, F4 = 433. Ultrasound-identified cirrhosis was noted in 283 patients. Using ultrasound-identified cirrhosis to predict compensated cirrhosis, the sensitivity was 34.0%, the specificity was 97.1%, the positive predictive value was 89.8%, the negative predictive value was 66.1%, the positive likelihood ratio was 11.6, and the negative likelihood ratio was 0.68. The area under the ROC curve (AUROC) was 0.66.Despite being affected by inter-observer variability, ultrasound is highly specific in diagnosing compensated cirrhosis in CHC patients in real world clinical practice. However, the sensitivity is low.

Diabetes is associated with advanced fibrosis and fibrosis progression in non-genotype 3 chronic hepatitis C patients

BACKGROUND

Diabetes is a risk factor of fibrosis progression in chronic hepatitis C (CHC). However, only one longitudinal study exploring whether diabetes is associated with progression from non-cirrhotic liver to cirrhosis in CHC patients has been conducted.

AIMS

We investigated whether diabetes is associated with progression from non-cirrhotic liver to cirrhosis in non-genotype 3 CHC patients.

METHODS

A cohort consisting of 976 non-genotype 3 patients histologically proven to have CHC was studied. After excluding patients with biopsy-proven or ultrasound-identified cirrhosis, there were 684 patients without cirrhosis. All 684 patients underwent hepatocellular carcinoma surveillance using ultrasound every 6 months, with a median duration of follow-up evaluation of 102.4 months. During the follow-up period, 60 patients developed cirrhosis according to ultrasound findings.

RESULTS

For the subgroup of 684 patients without cirrhosis, Kaplan-Meier survival analyses showed no significantly different cumulative incidences of cirrhosis (log-rank test; P = 0.71) among the patients with diabetes as compared to those without. However, after making adjustments for age, gender, fibrosis, steatosis, sustained virological response status, and obesity using Cox's proportional hazard model, diabetes was found to be an independent predictor for cirrhosis (HR = 1.9; 95% CI = 1.05-3.43, P = 0.03).

CONCLUSIONS

Diabetes is associated with progression from non-cirrhotic liver to cirrhosis in non-genotype 3 CHC patients.

Computer-aided Characterization and Diagnosis of Diffuse Liver Diseases Based on Ultrasound Imaging: A Review

Diffuse liver diseases, such as hepatitis, fatty liver, and cirrhosis, are becoming a leading cause of fatality and disability all over the world. Early detection and diagnosis of these diseases is extremely important to save lives and improve effectiveness of treatment. Ultrasound imaging, a noninvasive diagnostic technique, is the most commonly used modality for examining liver abnormalities. However, the accuracy of ultrasound-based diagnosis depends highly on expertise of radiologists. Computer-aided diagnosis systems based on ultrasound imaging assist in fast diagnosis, provide a reliable "second opinion" for experts, and act as an effective tool to measure response of treatment on patients undergoing clinical trials. In this review, we first describe appearance of liver abnormalities in ultrasound images and state the practical issues encountered in characterization of diffuse liver diseases that can be addressed by software algorithms. We then discuss computer-aided diagnosis in general with features and classifiers relevant to diffuse liver diseases. In later sections of this paper, we review the published studies and describe the key findings of those studies. A concise tabular summary comparing image database, features extraction, feature selection, and classification algorithms presented in the published studies is also exhibited. Finally, we conclude with a summary of key findings and directions for further improvements in the areas of accuracy and objectiveness of computer-aided diagnosis.

Cost-effectiveness of non-invasive methods for assessment and monitoring of liver fibrosis and cirrhosis in patients with chronic liver disease: systematic review and economic evaluation

BACKGROUND

Liver biopsy is the reference standard for diagnosing the extent of fibrosis in chronic liver disease; however, it is invasive, with the potential for serious complications. Alternatives to biopsy include non-invasive liver tests (NILTs); however, the cost-effectiveness of these needs to be established.

OBJECTIVE

To assess the diagnostic accuracy and cost-effectiveness of NILTs in patients with chronic liver disease.

DATA SOURCES

We searched various databases from 1998 to April 2012, recent conference proceedings and reference lists.

METHODS

We included studies that assessed the diagnostic accuracy of NILTs using liver biopsy as the reference standard. Diagnostic studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. Meta-analysis was conducted using the bivariate random-effects model with correlation between sensitivity and specificity (whenever possible). Decision models were used to evaluate the cost-effectiveness of the NILTs. Expected costs were estimated using a NHS perspective and health outcomes were measured as quality-adjusted life-years (QALYs). Markov models were developed to estimate long-term costs and QALYs following testing, and antiviral treatment where indicated, for chronic hepatitis B (HBV) and chronic hepatitis C (HCV). NILTs were compared with each other, sequential testing strategies, biopsy and strategies including no testing. For alcoholic liver disease (ALD), we assessed the cost-effectiveness of NILTs in the context of potentially increasing abstinence from alcohol. Owing to a lack of data and treatments specifically for fibrosis in patients with non-alcoholic fatty liver disease (NAFLD), the analysis was limited to an incremental cost per correct diagnosis. An analysis of NILTs to identify patients with cirrhosis for increased monitoring was also conducted.

RESULTS

Given a cost-effectiveness threshold of £20,000 per QALY, treating everyone with HCV without prior testing was cost-effective with an incremental cost-effectiveness ratio (ICER) of £9204. This was robust in most sensitivity analyses but sensitive to the extent of treatment benefit for patients with mild fibrosis. For HBV [hepatitis B e antigen (HBeAg)-negative)] this strategy had an ICER of £28,137, which was cost-effective only if the upper bound of the standard UK cost-effectiveness threshold range (£30,000) is acceptable. For HBeAg-positive disease, two NILTs applied sequentially (hyaluronic acid and magnetic resonance elastography) were cost-effective at a £20,000 threshold (ICER: £19,612); however, the results were highly uncertain, with several test strategies having similar expected outcomes and costs. For patients with ALD, liver biopsy was the cost-effective strategy, with an ICER of £822.

LIMITATIONS

A substantial number of tests had only one study from which diagnostic accuracy was derived; therefore, there is a high risk of bias. Most NILTs did not have validated cut-offs for diagnosis of specific fibrosis stages. The findings of the ALD model were dependent on assuptions about abstinence rates assumptions and the modelling approach for NAFLD was hindered by the lack of evidence on clinically effective treatments.

CONCLUSIONS

Treating everyone without NILTs is cost-effective for patients with HCV, but only for HBeAg-negative if the higher cost-effectiveness threshold is appropriate. For HBeAg-positive, two NILTs applied sequentially were cost-effective but highly uncertain. Further evidence for treatment effectiveness is required for ALD and NAFLD.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42011001561.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.

Prediction of liver cirrhosis, using diagnostic imaging tools

Early diagnosis of liver cirrhosis is important. Ultrasound-guided liver biopsy is the gold standard for diagnosis of liver cirrhosis. However, its invasiveness and sampling bias limit the applicability of the method. Basic imaging for the diagnosis of liver cirrhosis has developed over the last few decades, enabling early detection of morphological changes of the liver by ultrasonography (US), computed tomography, and magnetic resonance imaging (MRI). They are also accurate diagnostic methods for advanced liver cirrhosis, for which early diagnosis is difficult. There are a number of ways to compensate for this difficulty, including texture analysis to more closely identify the homogeneity of hepatic parenchyma, elastography to measure the stiffness and elasticity of the liver, and perfusion studies to determine the blood flow volume, transit time, and velocity. Amongst these methods, elastography using US and MRI was found to be slightly easier, faster, and able to provide an accurate diagnosis. Early diagnosis of liver cirrhosis using MRI or US elastography is therefore a realistic alternative, but further research is still needed.

Computer-aided assessment of hepatic contour abnormalities as an imaging biomarker for the prediction of hepatocellular carcinoma development in patients with chronic hepatitis C

PURPOSE

To evaluate whether a hepatic fibrosis index (HFI), quantified on the basis of hepatic contour abnormality, is a risk factor for the development of hepatocellular carcinoma (HCC) in patients with chronic hepatitis C.

MATERIALS AND METHODS

Our institutional review board approved this retrospective study and written informed consent was waved. During a 14-month period, consecutive 98 patients with chronic hepatitis C who had no medical history of HCC treatment (56 men and 42 women; mean age, 70.7 years; range, 48-91 years) were included in this study. Gadoxetic acid-enhanced hepatocyte specific phase was used to detect and analyze hepatic contour abnormality. Hepatic contour abnormality was quantified and converted to HFI using in-house proto-type software. We compared HFI between patients with (n=54) and without HCC (n=44). Serum levels of albumin, total bilirubin, aspartate transferase, alanine transferase, percent prothrombin time, platelet count, alpha-fetoprotein, protein induced by vitamin K absence-II, and HFI were tested as possible risk factors for the development of HCC by determining the odds ratio with logistic regression analysis.

RESULTS

HFIs were significantly higher in patients with HCC (0.58±0.86) than those without (0.36±0.11) (P<0.001). Logistic analysis revealed that only HFI was a significant risk factor for HCC development with an odds ratio (95% confidence interval) of 26.4 (9.0-77.8) using a cutoff value of 0.395.

CONCLUSION

The hepatic fibrosis index, generated using a computer-aided assessment of hepatic contour abnormality, may be a useful imaging biomarker for the prediction of HCC development in patients with chronic hepatitis C.

Accuracy of ultrasound to identify chronic liver disease

AIM: To identify and assess studies reporting the diagnostic performance of ultrasound imaging for identifying chronic liver disease (CLD) in a high risk population.

METHODS: A search was performed to identify studies investigating the diagnostic accuracy of ultrasound imaging for CLD. Two authors independently used the quality assessment of diagnostic accuracy studies (QUADAS) checklist to assess the methodological quality of the selected studies. Inter-observer reliability of the QUADAS tool was assessed by measuring the degree of agreement (percent agreement, κ statistic) between the reviewers for each assessment prior to a consensus meeting. The characteristics of each study population, sensitivity and specificity results for the index tests, and results of any testing for observer agreement were extracted from the reports. Receiver Operator Characteristic plots were generated using Microsoft Excel 2003 software and used to graphically display the diagnostic performance data and to explore the relationships between the reported ultrasound techniques and study characteristics, and methodology quality.

RESULTS: Twenty-one studies published between 1991 and 2009 were retained for data extraction, analysis and assessment for methodological quality. Assessment of methodology quality was performed on the 21 selected studies by two independent reviewers (RA & KT) using the QUADAS assessment tool. Across all studies the mean number of responses within the QUADAS assessment tool was 10 (range 7-13) for “Yes”, 1 (range 0-3) for “No” and 3 (range 0-6) for “unclear”. Inter-rater agreement for assessment of methodology quality was significantly greater than chance when assessing for representative spectrum, clear selection criteria, appropriate delay between reference and index tests, adequate descriptions of the index and reference tests, reference and index test blinding, and if relevant clinical information was provided. Seven studies reported moderate to high observer agreement for ultrasound techniques. Studies which clearly reported blinding performed better than the other studies for diagnostic accuracy, and lower diagnostic accuracy was evident for populations with lower prevalence of disease. Assessment of the liver surface using ultrasound consistently had moderate diagnostic accuracy across studies which demonstrated good research methodology. Other techniques demonstrated variable or poor to fair diagnostic accuracy.

CONCLUSION: Ultrasound of the liver surface is a useful diagnostic tool in patients at risk of CLD when assessing whether they should undergo a liver biopsy.