Artificial neural network-based models used for predicting 28- and 90-day mortality of patients with hepatitis B-associated acute-on-chronic liver failure

Hepatic encephalopathy post-TIPS: Current status and prospects in predictive assessment

Transjugular intrahepatic portosystemic shunt (TIPS) is an essential procedure for the treatment of portal hypertension but can result in hepatic encephalopathy (HE), a serious complication that worsens patient outcomes. Investigating predictors of HE after TIPS is essential to improve prognosis. This review analyzes risk factors and compares predictive models, weighing traditional scores such as Child-Pugh, Model for End-Stage Liver Disease (MELD), and albumin-bilirubin (ALBI) against emerging artificial intelligence (AI) techniques. While traditional scores provide initial insights into HE risk, they have limitations in dealing with clinical complexity. Advances in machine learning (ML), particularly when integrated with imaging and clinical data, offer refined assessments. These innovations suggest the potential for AI to significantly improve the prediction of post-TIPS HE. The study provides clinicians with a comprehensive overview of current prediction methods, while advocating for the integration of AI to increase the accuracy of post-TIPS HE assessments. By harnessing the power of AI, clinicians can better manage the risks associated with TIPS and tailor interventions to individual patient needs. Future research should therefore prioritize the development of advanced AI frameworks that can assimilate diverse data streams to support clinical decision-making. The goal is not only to more accurately predict HE, but also to improve overall patient care and quality of life.

Sarcopenia is associated with short- and long-term mortality in patients with acute-on-chronic liver failure

BACKGROUND

While sarcopenia is recognized as a predictor of mortality in cirrhosis, its influence on acute-on-chronic liver failure (ACLF) remains uncertain. Despite multiple studies examining the impact of sarcopenia on short-term mortality in patients with ACLF, the sample size of these studies was limited, and their outcomes were inconsistent. Therefore, this study aimed to explore the impact of sarcopenia on both short- and long-term mortality in patients with ACLF.

METHODS

This retrospective cohort study included 414 patients with ACLF that were treated between January 2016 and September 2022. Sarcopenia was diagnosed based on the measurement of the skeletal muscle index at the third lumbar vertebra (L3-SMI). Subsequently, the patients were divided into sarcopenia and non-sarcopenia groups. We analysed the basic clinical data of the two groups. Multivariate Cox proportional analysis was used to analyse short-term (28 days) and long-term (1 year and overall) mortality rates.

RESULTS

A total of 414 patients were included, with a mean age of 52.88 ± 13.41 years. Among them, 318 (76.8%) were male, and 239 (57.7%) had sarcopenia. A total of 280 (67.6%) patients died during the study period. Among them, 153 patients died within 28 days (37%) and 209 patients died within 1 year (50.5%). We found that the 28-day, 1-year and overall mortality rates in the sarcopenia group were significantly higher than those in the non-sarcopenia group (37% vs. 22.3%, P < 0.01; 50.5% vs. 34.9%, P < 0.01; and 67.6% vs. 53.1%, P < 0.01, respectively). Multivariate Cox regression analysis revealed that sarcopenia was significantly associated with increased mortality. The hazard ratios for sarcopenia were 2.05 (95% confidence interval [CI] 1.41-3.00, P < 0.01) for 28-day mortality, 1.81 (95% CI 1.29-2.54, P < 0.01) for 1-year mortality and 1.82 (95% CI 1.30-2.55, P < 0.01) for overall mortality. In addition, muscle density and international normalized ratio were associated with short- and long-term mortality.

CONCLUSIONS

Sarcopenia is associated with both short- and long-term mortality in patients with ACLF. Therefore, regular monitoring for sarcopenia is important for these patients.

From prognostication to therapeutics: Four key questions to accelerate the development of therapeutics for patients with acute-on-chronic liver failure

Acute-on-chronic liver failure (ACLF) has come a long way as a clinical concept within the hepatology and liver transplant communities. Though the term was proposed in 1995, the first recognition of the entity along with a consensus definition emerged in 2009. Subsequently, the entity has sparked great interest, inspired several consensus conferences, and inspired national societies to form professional ACLF affinity groups (eg, special interest group). Multicenter consortia have been established all over the world to study this condition, including the North American Consortium for the Study of End-Stage Liver Disease, Chronic Liver Failure consortium, Asian Pacific Association for the Study of Liver Diseases ACLF Research Consortium, Chronic Liver disease Evolution And Registry for Events and Decompensation, and the LiverHope Consortium. Collectively, these consortia have enrolled tens of thousands of patients with or at risk for ACLF across dozens of countries and characterized in detail the predictors, pathogenesis, and progression of patients with ACLF. Perhaps most importantly, they have produced essential data characterizing the excess morbidity and mortality that patients with ACLF face, making a compelling case for the urgent need for therapeutic strategies for this condition.

Predicting the 28-day prognosis of acute-on-chronic liver failure patients based on machine learning

BACKGROUND

We aimed to establish a prognostic predictive model based on machine learning (ML) methods to predict the 28-day mortality of acute-on-chronic liver failure (ACLF) patients, and to evaluate treatment effectiveness.

METHODS

ACLF patients from six tertiary hospitals were included for analysis. Features for ML models' development were selected by LASSO regression. Models' performance was evaluated by area under the curve (AUC) and accuracy. Shapley additive explanation was used to explain the ML model.

RESULTS

Of the 736 included patients, 587 were assigned to a training set and 149 to an external validation set. Features selected included age, hepatic encephalopathy, total bilirubin, PTA, and creatinine. The eXtreme Gradient Boosting (XGB) model outperformed other ML models in the prognostic prediction of ACLF patients, with the highest AUC and accuracy. Delong's test demonstrated that the XGB model outperformed Child-Pugh score, MELD score, CLIF-SOFA, CLIF-C OF, and CLIF-C ACLF. Sequential assessments at baseline, day 3, day 7, and day 14 improved the predictive performance of the XGB-ML model and can help clinicians evaluate the effectiveness of medical treatment.

CONCLUSIONS

We established an XGB-ML model to predict the 28-day mortality of ACLF patients as well as to evaluate the treatment effectiveness.

The value of dynamic changes in FT3 level for predicting 90-day prognosis of HBV-ACLF patients

OBJECTIVE

To explore the effect of dynamic changes in free triiodothyronine (FT3) level for predicting the 90 day prognosis of patients with hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF).

METHODS

The clinical data of 122 hospitalised patients with HBV-ACLF between September 2018 and January 2020 were collected and divided into a survival group (77 cases) and a death group (45 cases) according to the 90 day prognosis. We statistically analysed the characteristics of FT3 changes in the two groups of patients. Binary logistic regression one-way analysis was used to assess the degree of influence of each factor. The Kaplan-Meier survival curve and receiver operating characteristic curve were used to evaluate the effect of a single change in FT3 level difference (single △FT3) and the FT3 level change range (△FT3 range) in predicting the 90-day prognosis of patients.

RESULTS

There were only three types of changes in FT3 levels, which included 19 (15.6%) cases of continuous normal type, 35 (28.7%) cases of continuous decrease type and 68 (55.7%) cases of U-shaped change type. The difference in survival curves between the three types of patients was statistically significant (P < 0.001).

CONCLUSION

The dynamic change type of FT3 is related to the disease severity and 90-day prognosis of patients with HBV-ACLF. The single FT3 value and FT3 range could be used as a predictive factor for the 90-day prognosis of patients with HBV-ACLF. These results have a degree of research value and are worth further exploration in the future.

Machine Learning for Short-Term Mortality in Acute Decompensation of Liver Cirrhosis: Better than MELD Score

Prediction of short-term mortality in patients with acute decompensation of liver cirrhosis could be improved. We aimed to develop and validate two machine learning (ML) models for predicting 28-day and 90-day mortality in patients hospitalized with acute decompensated liver cirrhosis. We trained two artificial neural network (ANN)-based ML models using a training sample of 165 out of 290 (56.9%) patients, and then tested their predictive performance against Model of End-stage Liver Disease-Sodium (MELD-Na) and MELD 3.0 scores using a different validation sample of 125 out of 290 (43.1%) patients. The area under the ROC curve (AUC) for predicting 28-day mortality for the ML model was 0.811 (95%CI: 0.714- 0.907; p < 0.001), while the AUC for the MELD-Na score was 0.577 (95%CI: 0.435-0.720; p = 0.226) and for MELD 3.0 was 0.600 (95%CI: 0.462-0.739; p = 0.117). The area under the ROC curve (AUC) for predicting 90-day mortality for the ML model was 0.839 (95%CI: 0.776- 0.884; p < 0.001), while the AUC for the MELD-Na score was 0.682 (95%CI: 0.575-0.790; p = 0.002) and for MELD 3.0 was 0.703 (95%CI: 0.590-0.816; p < 0.001). Our study demonstrates that ML-based models for predicting short-term mortality in patients with acute decompensation of liver cirrhosis perform significantly better than MELD-Na and MELD 3.0 scores in a validation cohort.

Prognostic value of neutrophil count to albumin ratio in patients with decompensated cirrhosis

Our study aimed to investigate the prognostic value of neutrophil count to albumin ratio (NAR) in predicting short-term mortality of patients with decompensated cirrhosis (DC). A total of 623 DC patients were recruited from a retrospective observational cohort study. They were admitted to our hospital from January 2014 to December 2015. NAR of each patient was calculated and analyzed for the association with 90-day liver transplantation-free (LT-free) outcome. The performance of NAR and the integrated model were tested by a receiver-operator curve (ROC) and C-index. The 90-day LT-free mortality of patients with DC was 10.6%. NAR was significantly higher in 90-day non-survivors than in survivors (The median: 1.73 vs 0.76, P < 0.001). A threshold of 1.40 of NAR differentiated patients with a high risk of death (27.45%) from those with a low risk (5.11%). By multivariate analysis, high NAR was independently associated with poor short-term prognosis (high group: 5.07 (2.78, 9.22)). NAR alone had an area under the ROC curve of 0.794 and C-index of 0.7789 (0.7287, 0.8291) in predicting 90-day mortality. The integrated MELD-NAR (iMELD) model had a higher area under the ROC (0.872) and C-index (0.8558 (0.8122, 0.8994)) than the original MELD in predicting 90-day mortality. NAR can be used as an independent predictor of poor outcomes for patients with DC during short-term follow-up.

Risk factors for the mortality of hepatitis B virus-associated acute-on-chronic liver failure: a systematic review and meta-analysis

BACKGROUND

Hepatitis B virus-associated acute-on-chronic liver failure (HBV-ACLF) has been confirmed as a prevalent form of end-stage liver disease in people subjected to chronic HBV infection. However, there has been rare in-depth research on the risk factors for the mortality of HBV-ACLF. This study aimed at determining the risk factors for the mortality of HBV-ACLF.

METHODS

The relevant research was selected from four electronic databases that have been published as of August 2023. The existing research was reviewed in accordance with the inclusion and exclusion criteria. The level of quality of previous research was evaluated using the Newcastle-Ottawa scale. Moreover, a pooled estimate of the odds ratios (ORs) with their associated 95% confidence intervals (CIs) was provided through a meta-analysis. The data were combined, and the risk variables that at least two studies had considered were analyzed. The publication bias was examined through Egger's test and Begg's test.

RESULTS

Twenty two studies that conformed to the inclusion criteria were selected from 560 trials. Eight risk variables in terms of HBV-ACLF mortality were determined, which covered INR (OR = 1.923, 95% CI = 1.664-2.221, P < 0.001), Monocytes (OR = 1.201, 95% CI = 1.113-1.296, P < 0.001), Cirrhosis (OR = 1.432, 95% CI = 1.210-1.696, P < 0.001), HE (OR = 2.553, 95% CI = 1.968-3.312, P < 0.001), HE grade (OR = 2.059, 95% CI = 1.561-2.717, P < 0.001), SBP (OR = 1.383, 95% CI = 1.080-1.769, P = 0.010), Hyponatremia (OR = 1.941, 95% CI = 1.614-2.334, P < 0.001), as well as HRS (OR = 2.610, 95% CI = 1.669-4.080, P < 0.001).

CONCLUSION

The most significant risk factors for HBV-ACLF mortality comprise HRS, HE, and HE grade, followed by INR and hyponatremia. The Monocytes, cirrhosis, and SBP have been confirmed as the additional key risk factors for HBV-ACLF mortality.

Acute on chronic liver failure: prognostic models and artificial intelligence applications

Critically ill patients presenting with acute on chronic liver failure (ACLF) represent a particularly vulnerable population due to various considerations surrounding the syndrome definition, lack of robust prospective evaluation of outcomes, and allocation of resources such as organs for transplantation. Ninety-day mortality related to ACLF is high and patients who do leave the hospital are frequently readmitted. Artificial intelligence (AI), which encompasses various classical and modern machine learning techniques, natural language processing, and other methods of predictive, prognostic, probabilistic, and simulation modeling, has emerged as an effective tool in various areas of healthcare. These methods are now being leveraged to potentially minimize physician and provider cognitive load and impact both short-term and long-term patient outcomes. However, the enthusiasm is tempered by ethical considerations and a current lack of proven benefits. In addition to prognostic applications, AI models can likely help improve the understanding of various mechanisms of morbidity and mortality in ACLF. Their overall impact on patient-centered outcomes and countless other aspects of patient care remains unclear. In this review, we discuss various AI approaches being utilized in healthcare and discuss the recent and expected future impact of AI on patients with ACLF through prognostic modeling and AI-based approaches.

Machine learning-based model for predicting the esophagogastric variceal bleeding risk in liver cirrhosis patients

BACKGROUND

Liver cirrhosis patients are at risk for esophagogastric variceal bleeding (EGVB). Herein, we aimed to estimate the EGVB risk in patients with liver cirrhosis using an artificial neural network (ANN).

METHODS

We included 999 liver cirrhosis patients hospitalized at the Beijing Ditan Hospital, Capital Medical University in the training cohort and 101 patients from Shuguang Hospital in the validation cohort. The factors independently affecting EGVB occurrence were determined via univariate analysis and used to develop an ANN model.

RESULTS

The 1-year cumulative EGVB incidence rates were 11.9 and 11.9% in the training and validation groups, respectively. A total of 12 independent risk factors, including gender, drinking and smoking history, decompensation, ascites, location and size of varices, alanine aminotransferase (ALT), γ-glutamyl transferase (GGT), hematocrit (HCT) and neutrophil-lymphocyte ratio (NLR) levels as well as red blood cell (RBC) count were evaluated and used to establish the ANN model, which estimated the 1-year EGVB risk. The ANN model had an area under the curve (AUC) of 0.959, which was significantly higher than the AUC for the North Italian Endoscopic Club (NIEC) (0.669) and revised North Italian Endoscopic Club (Rev-NIEC) indices (0.725) (all P <  0.001). Decision curve analyses revealed improved net benefits of the ANN compared to the NIEC and Rev-NIEC indices.

CONCLUSIONS

The ANN model accurately predicted the 1-year risk for EGVB in liver cirrhosis patients and might be used as a basis for risk-based EGVB surveillance strategies.

Research progress on short-term prognosis of acute-on-chronic liver failure

INTRODUCTION

Acute-on-chronic liver failure (ACLF) is a clinical syndrome characterized as a severe condition with rapid progression, poor therapeutic response and poor prognosis. Early and timely evaluation of the prognosis is helpful for providing appropriate clinical intervention and prolonging patient survival.

AREAS COVERED

Currently, there are no specific dynamic and comprehensive approaches to assess the prognosis of patients with ACLF. This article reviews the progress in evaluating the short-term prognosis of ACLF to provide future directions for more dynamic prospective large-scale multicenter studies and a basis for individualized and precise treatment for ACLF patients. We searched PubMed and Web of Science with the term 'acute on chronic liver failure' and 'prognosis.' There was no date or language restriction, and our final search was on 26 October 2022.

EXPERT OPINION

ACLF is a dynamic process, and the best prognostic marker is the clinical evolution of organ failure over time. New prognostic markers are developing not only in the fields of genetics and histology but also toward diversification combined with imaging. Determining which patients will benefit from continued advanced life support is a formidable challenge, and accurate short-term prognostic assessments of ACLF are a good approach to addressing this issue.

Risk Association of Liver Cancer and Hepatitis B with Tree Ensemble and Lifestyle Features

The second-largest cause of death by cancer in Korea is liver cancer, which leads to acute morbidity and mortality. Hepatitis B is the most common cause of liver cancer. About 70% of liver cancer patients suffer from hepatitis B. Early risk association of liver cancer and hepatitis B can help prevent fatal conditions. We propose a risk association method for liver cancer and hepatitis B with only lifestyle features. The diagnostic features were excluded to reduce the cost of gathering medical data. The data source is the Korea National Health and Nutrition Examination Survey (KNHANES) from 2007 to 2019. We use 3872 and 4640 subjects for liver cancer and hepatitis B model, respectively. Random forest is employed to determine functional relationships between liver diseases and lifestyle features. The performance of our proposed method was compared with six machine learning methods. The results showed the proposed method outperformed the other methods in the area under the receiver operator characteristic curve of 0.8367. The promising results confirm the superior performance of the proposed method and show that the proposed method with only lifestyle features provides significant advantages, potentially reducing the cost of detecting patients who require liver health care in advance.

Machine learning-based system for prediction of ascites grades in patients with liver cirrhosis using laboratory and clinical data: design and implementation study

OBJECTIVES

The aim of the study was to implement a non-invasive model to predict ascites grades among patients with cirrhosis.

METHODS

In the present study, we used modern machine learning (ML) methods to develop a scoring system solely based on routine laboratory and clinical data to help physicians accurately diagnose and predict different degrees of ascites. We used ANACONDA3-5.2.0 64 bit, free and open-source platform distribution of Python programming language with numerous modules, packages, and rich libraries that provide various methods for classification problems. Through the 10-fold cross-validation, we employed three common learning models on our dataset, k-nearest neighbors (KNN), support vector machine (SVM), and neural network classification algorithms.

RESULTS

According to the data received from the research institute, three types of data analysis have been performed. The algorithms used to predict ascites were KNN, cross-validation (CV), and multilayer perceptron neural networks (MLPNN), which achieved an average accuracy of 94, 91, and 90%, respectively. Also, in the average accuracy of the algorithms, KNN had the highest accuracy of 94%.

CONCLUSIONS

We applied well-known ML approaches to predict ascites. The findings showed a strong performance compared to the classical statistical approaches. This ML-based approach can help to avoid unnecessary risks and costs for patients with acute stages of the disease.

Application of artificial intelligence in non-alcoholic fatty liver disease and viral hepatitis

Non-alcoholic fatty liver disease (NAFLD) and chronic viral hepatitis are among the most significant causes of liver-related mortality worldwide. It is critical to develop reliable methods of predicting progression to fibrosis, cirrhosis, and decompensated liver disease. Current screening methods such as biopsy and transient elastography are limited by invasiveness and observer variation in analysis of data. Artificial intelligence (AI) provides a unique opportunity to more accurately diagnose NAFLD and viral hepatitis, and to identify patients at high risk for disease progression. We conducted a literature review of existing evidence for AI in NAFLD and viral hepatitis. Thirteen articles on AI in NAFLD and 14 on viral hepatitis were included in our analysis. We found that machine learning algorithms were comparable in accuracy to current methods for diagnosis and fibrosis prediction (MELD-Na score, liver biopsy, FIB-4 score, and biomarkers). They also reliably predicted hepatitis C treatment failure and hepatic encephalopathy, for which there are currently no established prediction tools. These studies show that AI could be a helpful adjunct to existing techniques for diagnosing, monitoring, and treating both NAFLD and viral hepatitis.

Recent Advances in Understanding, Diagnosing, and Treating Hepatitis B Virus Infection

Chronic hepatitis B virus (HBV) infection affects 292 million people worldwide and is associated with a broad range of clinical manifestations including cirrhosis, liver failure, and hepatocellular carcinoma (HCC). Despite the availability of an effective vaccine HBV still causes nearly 900,000 deaths every year. Current treatment options keep HBV under control, but they do not offer a cure as they cannot completely clear HBV from infected hepatocytes. The recent development of reliable cell culture systems allowed for a better understanding of the host and viral mechanisms affecting HBV replication and persistence. Recent advances into the understanding of HBV biology, new potential diagnostic markers of hepatitis B infection, as well as novel antivirals targeting different steps in the HBV replication cycle are summarized in this review article.