Diagnostic value of combined serum biomarkers for the evaluation of liver fibrosis in chronic hepatitis C infection: A multicenter, noninterventional, observational study

Integrated neural network and evolutionary algorithm approach for liver fibrosis staging: Can artificial intelligence reduce patient costs?

Background and Aim

Staging liver fibrosis is important, and liver biopsy is the gold standard diagnostic tool. We aim to design and evaluate an artificial neural network (ANN) method by taking advantage of the Teaching Learning-Based Optimization (TLBO) algorithm for the prediction of liver fibrosis stage in blood donors and hepatitis C patients.

Methods

We propose a method based on a selection of machine learning classification methods including multilayer perceptron (MLP) neural network, Naive Bayesian (NB), decision tree, and deep learning. Initially, the synthetic minority oversampling technique (SMOTE) is performed to address the imbalance in the dataset. Afterward, the integration of MLP and TLBO is implemented.

Results

We propose a novel algorithm that reduces the number of required patient features to seven inputs. The accuracy of MLP using 12 features is 0.903, while that of the proposed MLP with TLBO is 0.891. Besides, the diagnostic accuracy of all methods, except the model designed with the Bayesian network, increases when the SMOTE balancer is applied.

Conclusion

The decision tree-based deep learning methods show the highest levels of accuracy with 12 features. Interestingly, with the use of TLBO and seven features, MLP reached an accuracy rate of 0.891, which is quite satisfactory when compared with those of similar studies. The proposed model provides high diagnostic accuracy, while reducing the required number of properties from the samples. The results of our study show that the recruited algorithm of our study is more straightforward, with a smaller number of required properties and similar accuracy.

How Successful Are APRI and FIB-4 Scores in Predicting Liver Fibrosis in Chronic Hepatitis B Patients?

Objective

We aimed to evaluate the correlation of fibrosis severity in liver biopsies, the gold standard for the diagnosis of patients with chronic hepatitis B (CHB), using noninvasive methods such as the aspartate aminotransferase (AST)-to-platelet ratio index (APRI) and fibrosis-4 score (FIB-4).

Materials and Methods

The study included patients who were followed and treated for CHB in 2018-2023. Biochemical markers and liver biopsy findings of the cases were retrospectively, and their correlations with APRI and FIB-4, which are noninvasive scores, were compared.

Results

The study included 202 patients. The biochemical markers and liver biopsy findings of the cases were examined retrospectively, and their correlations with the noninvasive scores APRI and FIB-4 were compared. According to liver biopsy results, 109 (54.0%) cases had no fibrosis (stage 0.1), 85 (42.1%) cases had mild fibrosis (stage 2, 3), and 8 (4%) cases had severe fibrosis (stage 4, 5, 6). The median FIB-4 score was 0.79 (0.25 -11.74), and the median APRI score was 0.29 (0.10-29.40). When the predictive power of the APRI score to discriminate between "without fibrosis" and "with fibrosis (mild and severe)" was evaluated by receiver operating characteristic (ROC) curve analysis, for the APRI score >0.408 as the ideal cut-off point, the sensitivity and specificity were found to be 34% and 79%, respectively. When the cut-off point for the FIB-4 score was >0.701, the sensitivity and specificity were 71% and 46%, respectively. Although the area under the curve (AUC) ratios ranged between 52% and 64% in the ROC analyses, the sensitivity ratios of the cut-off points calculated for FIB-4 were higher. The likelihood ratios of the cut-off point we found for the APRI score (1.61 and 1.75, respectively) were relatively better than those for FIB-4 (1.31 and 1.41, respectively).

Conclusion

Noninvasive tests used to detect liver fibrosis in individuals with CHB do not eliminate the need for liver biopsy but may provide insight into the fibrosis status of patients.

Noninvasive Fibrosis Assessment in Chronic Hepatitis C Infection: An Update

Liver biopsy is historically the gold standard for liver fibrosis assessment of chronic hepatitis C patients. However, with the introduction and validation of noninvasive tests (NITs) to evaluate advanced fibrosis, and the direct-acting antiviral agents for treatment of chronic hepatitis C virus (HCV), the role of NITs have become even more complex. There is now need for longitudinal monitoring and elucidation of cutoff values for prediction of liver-related complication after sustained virological response. The aim of this report is to provide a critical overview of the various NITs available for the assessment of liver fibrosis in HCV patients.

Evaluation of fibrosis with noninvasive biochemical tests in chronic viral hepatitis B

Background and Aim

Early diagnosis and treatment of chronic hepatitis B (CHB) disease are important for the prevention of complications such as cirrhosis and hepatocellular cancer. Liver biopsy is an invasive, complicated, and expensive diagnostic method, which is the gold standard for detecting fibrosis. The aim of this study was to investigate the role of these tests in predicting liver fibrosis and treatment decision.

Materials and Methods

A total of 1051 patients diagnosed with CHB between 2010 and 2020 in the Gaziantep University Gastroenterology Department were retrospectively evaluated. AAR, API, APRI, FIB-4, KING score, and FIBROQ score were calculated at the time of onset diagnosis. In addition, the Zeugma score, a new formula that is thought to be more sensitive and specific, was determined. Noninvasive fibrosis scores were compared according to the biopsy results of the patients.

Results

In this study, the area values under the curve were 0.648 for the API score, 0.711 for the APRI score, 0.716 for the FIB-4 score, 0.723 for the KING score, 0.595 for the FIBROQ score, and 0.701 for the Zeugma score (p<0.05). No statistically significant difference was obtained for the AAR score. The KING, FIB-4, APRI, and Zeugma scores were the best indicators for detecting advanced fibrosis. For KING, FIB-4, APRI, and Zeugma scores, the cutoff value for the prediction of advanced fibrosis were ≥8.67, ≥0.94, ≥16.24, and ≥9.63 with a sensitivity of 50.52%, 56.77%, 59.64%, and 52.34%, specificity of 87.26%, 74.96%, 73.61%, and 78.11%, respectively (p<0.05). In our study, we compared the globulin and GGT parameters with fibrosis, which we used in the Zeugma score formula. Globulin and GGT mean values were significantly higher in the fibrosis group (p<0.05). There was a statistically significant correlation between fibrosis and globulin and GGT values (p<0.05, r=0.230 and p<0.05, r=0.305, respectively).

Conclusion

The KING score was found to be the most reliable method for the noninvasive detection of hepatic fibrosis in patients with chronic HBV. The FIB-4, APRI, and Zeugma scores were also shown to be effective in determining liver fibrosis. It was shown that the AAR score was not sufficient for detecting hepatic fibrosis. The Zeugma score, a novel noninvasive test, is a useful and easy tool to evaluate liver fibrosis in patients with chronic HBV and has better accuracy than AAR, API, and FIBROQ.

Clinical application of deep learning and radiomics in hepatic disease imaging: a systematic scoping review

OBJECTIVE

Artificial intelligence (AI) has begun to play a pivotal role in hepatic imaging. This systematic scoping review summarizes the latest progress of AI in evaluating hepatic diseases based on computed tomography (CT) and magnetic resonance (MR) imaging.

METHODS

We searched PubMed and Web of Science for publications, using terms related to deep learning, radiomics, imaging methods (CT or MR), and the liver. Two reviewers independently selected articles and extracted data from each eligible article. The Quality Assessment of Diagnostic Accuracy Studies-AI (QUADAS-AI) tool was used to assess the risk of bias and concerns regarding applicability.

RESULTS

The screening identified 45 high-quality publications from 235 candidates, including 8 on diffuse liver diseases and 37 on focal liver lesions. Nine studies used deep learning and 36 studies used radiomics. All 45 studies were rated as low risk of bias in patient selection and workflow, but 36 (80%) were rated as high risk of bias in the index test because they lacked external validation. In terms of concerns regarding applicability, all 45 studies were rated as low concerns. These studies demonstrated that deep learning and radiomics can evaluate liver fibrosis, cirrhosis, portal hypertension, and a series of complications caused by cirrhosis, predict the prognosis of malignant hepatic tumors, and differentiate focal hepatic lesions.

CONCLUSIONS

The latest studies have shown that deep learning and radiomics based on hepatic CT and MR imaging have potential application value in the diagnosis, treatment evaluation, and prognosis prediction of common liver diseases. The AI methods may become useful tools to support clinical decision-making in the future.

ADVANCES IN KNOWLEDGE

Deep learning and radiomics have shown their potential in the diagnosis, treatment evaluation, and prognosis prediction of a series of common diffuse liver diseases and focal liver lesions.

Clinical and novel application of FibroScan, FIB-4 and aspartate aminotransferase-to-platelet ratio index in liver fibrosis evaluation in patients with hepatocellular carcinoma and their roles in oesophageal variceal prediction

BACKGROUND

Non-invasive techniques for liver fibrosis diagnosis are very important for clinician especially in high-risk patients for liver biopsy. We further explored the diagnostic accuracy of FibroScan, FIB-4 and aminotransferase-to-platelet ratio index (APRI) in identifying liver fibrosis and assess their predictive role for oesophageal varices in patients with hepatocellular carcinoma (HCC).

METHODS

In total, 380 patients who underwent surgery for HCC were included based on retrospective study design. Liver fibrosis was pathologically diagnosed using the Ishak scoring system. Liver stiffness parameters were measured using FibroScan. APRI and FIB-4 were calculated. Among those, 121 patients who received oesophagogastroduodenoscopic examination underwent variceal evaluation.

RESULTS

For liver cirrhosis diagnosis with FibroScan, the optimal cut-off values for the patients with HCC overall, left HCC and right HCC were 8.85, 11.75 and 8.70 kPa (the accuracy were 78.7%, 78.4% and 79.2%, respectively). They had high areas under the receiver operating characteristic curve of 0.84, 0.84 and 0.85. The combined FibroScan, APRI and FIB-4 had very high specificity (more than 92%) for cirrhosis diagnosis. The optimal cut-off liver stiffness values for the diagnosis of varices were all 11.2 kPa. For predicting varices, the optimal cut-off values of FIB-4 and APRI were 2.64 and 0.71, their accuracy were 64.3%-78.4%, 69.4% and 72.7%, respectively.

CONCLUSIONS

FibroScan, FIB-4 and APRI have moderate accuracy for liver fibrosis diagnosis and oesophageal varices prediction in patients with hepatoma. This is a study of these non-invasive techniques applied in specific hepatoma patients and with inevitable limitations and need future more studies for validation.

Hepatic Steatosis Predicts Higher Incidence of Recurrence in Colorectal Cancer Liver Metastasis Patients

Purpose: Colorectal liver metastasis (CRLM) is the major cause of death due to colorectal cancer. Although great efforts have been made in treatment of CRLM, about 60–70% of patients will develop hepatic recurrence. Hepatic steatosis was reported to provide fertile soil for metastasis. However, whether hepatic steatosis predicts higher incidence of CRLM recurrence is not clear. Therefore, we aimed to determine the role of hepatic steatosis in CRLM recurrence in the present study.

Methods: Consecutive CRLM patients undergoing curative treatment were retrospectively enrolled and CT liver-spleen attenuation ratio was used to detect the presence of hepatic steatosis. In patients with hepatic steatosis, we also detected the presence of fibrosis. Besides, a systematic literature search was performed to do meta-analysis to further analyze the association between hepatic steatosis and CRLM recurrence.

Results: A total of 195 eligible patients were included in our center. Patients with hepatic steatosis had a significantly worse overall (P = 0.0049) and hepatic recurrence-free survival (RFS) (P = 0.0012). Univariate and multivariate analysis confirmed its essential role in prediction of RFS. Besides, hepatic fibrosis is associated with worse overall RFS (P = 0.039) and hepatic RFS (P = 0.048). In meta-analysis, we included other four studies, with a total of 1,370 patients in the case group, and 3,735 patients in the control group. The odds ratio was 1.98 (95% CI: 1.25–3.14, P = 0.004), indicating that patients with steatosis had a significantly higher incidence of CRLM recurrence.

Conclusion: In summary, patients with hepatic steatosis had a significantly worse overall and hepatic RFS and it's associated with higher incidence of CRLM recurrence.

Application of artificial intelligence in chronic liver diseases: a systematic review and meta-analysis

Background

The gold standard for the diagnosis of liver fibrosis and nonalcoholic fatty liver disease (NAFLD) is liver biopsy. Various noninvasive modalities, e.g., ultrasonography, elastography and clinical predictive scores, have been used as alternatives to liver biopsy, with limited performance. Recently, artificial intelligence (AI) models have been developed and integrated into noninvasive diagnostic tools to improve their performance.

Methods

We systematically searched for studies on AI-assisted diagnosis of liver fibrosis and NAFLD on MEDLINE, Scopus, Web of Science and Google Scholar. The pooled sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and diagnostic odds ratio (DOR) with their 95% confidence intervals (95% CIs) were calculated using a random effects model. A summary receiver operating characteristic curve and the area under the curve was generated to determine the diagnostic accuracy of the AI-assisted system. Subgroup analyses by diagnostic modalities, population and AI classifiers were performed.

Results

We included 19 studies reporting the performances of AI-assisted ultrasonography, elastrography, computed tomography, magnetic resonance imaging and clinical parameters for the diagnosis of liver fibrosis and steatosis. For the diagnosis of liver fibrosis, the pooled sensitivity, specificity, PPV, NPV and DOR were 0.78 (0.71–0.85), 0.89 (0.81–0.94), 0.72 (0.58–0.83), 0.92 (0.88–0.94) and 31.58 (11.84–84.25), respectively, for cirrhosis; 0.86 (0.80–0.90), 0.87 (0.80–0.92), 0.85 (0.75–0.91), 0.88 (0.82–0.92) and 37.79 (16.01–89.19), respectively; for advanced fibrosis; and 0.86 (0.78–0.92), 0.81 (0.77–0.84), 0.88 (0.80–0.93), 0.77 (0.58–0.89) and 26.79 (14.47–49.62), respectively, for significant fibrosis. Subgroup analyses showed significant differences in performance for the diagnosis of fibrosis among different modalities. The pooled sensitivity, specificity, PPV, NPV and DOR were 0.97 (0.76–1.00), 0.91 (0.78–0.97), 0.95 (0.87–0.98), 0.93 (0.80–0.98) and 191.52 (38.82–944.81), respectively, for the diagnosis of liver steatosis.

Conclusions

AI-assisted systems have promising potential for the diagnosis of liver fibrosis and NAFLD. Validations of their performances are warranted before implementing these AI-assisted systems in clinical practice.

Trial registration: The protocol was registered with PROSPERO (CRD42020183295).

Aspartate aminotransferase-to-platelet ratio index (APRI): A potential marker for diagnosis in patients at risk of severe malaria caused by Plasmodium vivax

Acute infection with Plasmodium vivax, classically associated with benign disease, has been presenting as serious and even fatal disease in recent years. Severe disease is mainly due to biochemical and hematological alterations during the acute phase of infection. In the present cross-sectional study, the aspartate aminotransferase-to-platelet ratio index (APRI) was evaluated as a method for identifying patients at risk of severe vivax malaria. This retrospective study included 130 patients with confirmed P. vivax infection between June 2006 and January 2018. Clinical-epidemiological data were obtained from medical records. Hematological and biochemical parameters were determined using automated equipment. The criteria of severity for infection by Plasmodium falciparum, established by the World Health Organization (WHO), were adapted to classify patients with danger signs of severe vivax malaria. Of the 130 patient’s records evaluated, 19 (14.6%) had one or more signs and symptoms of severe malaria. The mean APRI values among patients with and without severe malaria were 2.11 and 1.09, respectively (p = 0.044). Among those with severe disease, the proportion with an APRI value above 1.50 was 30% compared to the 10% among those without severe disease (p = 0.007). The area under the receiver operating characteristic curve (95% CI), calculated to assess the accuracy of the APRI in discriminating between patients with and without severe disease, was 0.645 (0.494; 0.795). An APRI cutoff of 0.74 resulted in sensitivity of 74.0%, specificity of 56.0%, and accuracy of 65.0%. This study shows that the APRI is elevated in patients with evidence of infection by P. vivax. Based on the good sensitivity found in this study, we conclude that this simple index can serve as a diagnostic biomarker to identify patients at risk of severe disease during the acute phase of P. vivax infection.