Laboratory parameter-based machine learning model for excluding non-alcoholic fatty liver disease (NAFLD) in the general population

Application of artificial intelligence in liver diseases: From diagnosis to treatment

Infectious or noninfectious liver disease has inexorably risen as one of the leading causes of global death and disease burden. There were an estimated 2.14 million liver-related deaths in 2017, representing an 11.4% increase since 2012. Traditional diagnosis and treatment methods have various dilemmas in different causes of liver disease. As a hot research topic in recent years, the application of artificial intelligence (AI) in different fields has attracted extensive attention, and new technologies have brought more ideas for the diagnosis and treatment of some liver diseases. Machine learning (ML) is the core of AI and the basic way to make a computer intelligent. ML technology has many potential uses in hepatology, ranging from exploring new noninvasive means to predict or diagnose different liver diseases to automated image analysis. The application of ML in liver diseases can help clinical staff to diagnose and treat different liver diseases quickly, accurately and scientifically, which is of importance for reducing the incidence and mortality of liver diseases, reducing medical errors, and promoting the development of medicine. This paper reviews the application and prospects of AI in liver diseases, and aims to improve clinicians’ awareness of the importance of AI in the diagnosis and treatment of liver diseases.

Artificial intelligence in gastroenterology: A state-of-the-art review

The development of artificial intelligence (AI) has increased dramatically in the last 20 years, with clinical applications progressively being explored for most of the medical specialties. The field of gastroenterology and hepatology, substantially reliant on vast amounts of imaging studies, is not an exception. The clinical applications of AI systems in this field include the identification of premalignant or malignant lesions (e.g., identification of dysplasia or esophageal adenocarcinoma in Barrett’s esophagus, pancreatic malignancies), detection of lesions (e.g., polyp identification and classification, small-bowel bleeding lesion on capsule endoscopy, pancreatic cystic lesions), development of objective scoring systems for risk stratification, predicting disease prognosis or treatment response [e.g., determining survival in patients post-resection of hepatocellular carcinoma), determining which patients with inflammatory bowel disease (IBD) will benefit from biologic therapy], or evaluation of metrics such as bowel preparation score or quality of endoscopic examination. The objective of this comprehensive review is to analyze the available AI-related studies pertaining to the entirety of the gastrointestinal tract, including the upper, middle and lower tracts; IBD; the hepatobiliary system; and the pancreas, discussing the findings and clinical applications, as well as outlining the current limitations and future directions in this field.

Machine learning models for predicting non-alcoholic fatty liver disease in the general United States population: NHANES database

BACKGROUND

Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease, affecting over 30% of the United States population. Early patient identification using a simple method is highly desirable.

AIM

To create machine learning models for predicting NAFLD in the general United States population.

METHODS

Using the NHANES 1988-1994. Thirty NAFLD-related factors were included. The dataset was divided into the training (70%) and testing (30%) datasets. Twenty-four machine learning algorithms were applied to the training dataset. The best-performing models and another interpretable model (i.e., coarse trees) were tested using the testing dataset.

RESULTS

There were 3235 participants (n = 3235) that met the inclusion criteria. In the training phase, the ensemble of random undersampling (RUS) boosted trees had the highest F1 (0.53). In the testing phase, we compared selective machine learning models and NAFLD indices. Based on F1, the ensemble of RUS boosted trees remained the top performer (accuracy 71.1% and F1 0.56) followed by the fatty liver index (accuracy 68.8% and F1 0.52). A simple model (coarse trees) had an accuracy of 74.9% and an F1 of 0.33.

CONCLUSION

Not every machine learning model is complex. Using a simpler model such as coarse trees, we can create an interpretable model for predicting NAFLD with only two predictors: fasting C-peptide and waist circumference. Although the simpler model does not have the best performance, its simplicity is useful in clinical practice.

Development and validation of a neural network for NAFLD diagnosis

Non-Alcoholic Fatty Liver Disease (NAFLD) affects about 20–30% of the adult population in developed countries and is an increasingly important cause of hepatocellular carcinoma. Liver ultrasound (US) is widely used as a noninvasive method to diagnose NAFLD. However, the intensive use of US is not cost-effective and increases the burden on the healthcare system. Electronic medical records facilitate large-scale epidemiological studies and, existing NAFLD scores often require clinical and anthropometric parameters that may not be captured in those databases. Our goal was to develop and validate a simple Neural Network (NN)-based web app that could be used to predict NAFLD particularly its absence. The study included 2970 subjects; training and testing of the neural network using a train–test-split approach was done on 2869 of them. From another population consisting of 2301 subjects, a further 100 subjects were randomly extracted to test the web app. A search was made to find the best parameters for the NN and then this NN was exported for incorporation into a local web app. The percentage of accuracy, area under the ROC curve, confusion matrix, Positive (PPV) and Negative Predicted Value (NPV) values, precision, recall and f1-score were verified. After that, Explainability (XAI) was analyzed to understand the diagnostic reasoning of the NN. Finally, in the local web app, the specificity and sensitivity values were checked. The NN achieved a percentage of accuracy during testing of 77.0%, with an area under the ROC curve value of 0.82. Thus, in the web app the NN evidenced to achieve good results, with a specificity of 1.00 and sensitivity of 0.73. The described approach can be used to support NAFLD diagnosis, reducing healthcare costs. The NN-based web app is easy to apply and the required parameters are easily found in healthcare databases.

Challenges and opportunities in the application of artificial intelligence in gastroenterology and hepatology

Artificial intelligence (AI) is an umbrella term used to describe a cluster of interrelated fields. Machine learning (ML) refers to a model that learns from past data to predict future data. Medicine and particularly gastroenterology and hepatology, are data-rich fields with extensive data repositories, and therefore fruitful ground for AI/ML-based software applications. In this study, we comprehensively review the current applications of AI/ML-based models in these fields and the opportunities that arise from their application. Specifically, we refer to the applications of AI/ML-based models in prevention, diagnosis, management, and prognosis of gastrointestinal bleeding, inflammatory bowel diseases, gastrointestinal premalignant and malignant lesions, other nonmalignant gastrointestinal lesions and diseases, hepatitis B and C infection, chronic liver diseases, hepatocellular carcinoma, cholangiocarcinoma, and primary sclerosing cholangitis. At the same time, we identify the major challenges that restrain the widespread use of these models in healthcare in an effort to explore ways to overcome them. Notably, we elaborate on the concerns regarding intrinsic biases, data protection, cybersecurity, intellectual property, liability, ethical challenges, and transparency. Even at a slower pace than anticipated, AI is infiltrating the healthcare industry. AI in healthcare will become a reality, and every physician will have to engage with it by necessity.

Novel artificial intelligent/neural network system for staging of nonalcoholic steatohepatitis

AIM

To develop a novel noninvasive test using an artificial intelligence (AI)/neural network (NN) system (named Fibro-Scope) to determine the fibrosis stage in nonalcoholic steatohepatitis (NASH).

METHODS

Three hundred twenty-four and 110 patients with histologically diagnosed nonalcoholic fatty liver disease (NAFLD) were enrolled for training and validation studies, respectively. Two independent pathologists histologically diagnosed patients with NAFLD for the validation study. Fibro-Scope was undertaken using 12 items: age, sex, height, weight, waist circumference, aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transferase, cholesterol, triglyceride, platelet count, and type 4 collagen 7s.

RESULTS

Differentiation of F0 versus F1-4 using the Fibro-Scope revealed 99.5% sensitivity, 90.9% specificity, 97.4% positive predictive value, and 98.0% negative predictive value in a training study with gray zone analysis, which was also effective in the analysis without gray zone. Discrimination was also excellent when comparing F0-1 versus F2-4 and F0-2 versus F3-4. In a validation study with gray zone analysis, differentiation of F0 from F1-4 using Fibro-Scope was also excellent. The discrimination of F0-1 from F2-4 using Fibro-Scope with gray zone analysis showed over 80% sensitivity and specificity in the histological diagnosis of both pathologists, but was lower without the gray zone analysis. The discrimination of F0-2 from F3-4 was effective in the analysis with gray zone; however, their sensitivity and specificity were slightly inferior in the analysis without gray zone.

CONCLUSIONS

Artificial intelligence/neural network algorithms termed Fibro-Scope are easy to use and can accurately differentially diagnose minimal, moderate, and advanced fibrosis. Fibro-Scope will promote rapid NASH diagnosis and facilitate diagnosing the fibrosis stage in NASH.

Preparing for the NASH epidemic: A call to action

Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) are common conditions with a rising burden. Yet there are significant management gaps between clinical guidelines and practice in patients with NAFLD and NASH. Further, there is no single global guiding strategy for the management of NAFLD and NASH. The American Gastroenterological Association, in collaboration with 7 professional associations, convened an international conference comprising 32 experts in gastroenterology, hepatology, endocrinology, and primary care providers from the United States, Europe, Asia, and Australia. Conference content was informed by the results of a national NASH Needs Assessment Survey. The participants reviewed and discussed published literature on global burden, screening, risk stratification, diagnosis, and management of individuals with NAFLD, including those with NASH. Participants identified promising approaches for clinical practice and prepared a comprehensive, unified strategy for primary care providers and relevant specialists encompassing the full spectrum of NAFLD/NASH care. They also identified specific high-yield targets for clinical research and called for a unified, international public health response to NAFLD and NASH.

Preparing for the NASH Epidemic: A Call to Action

Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) are common conditions with a rising burden. Yet there are significant management gaps between clinical guidelines and practice in patients with NAFLD and NASH. Further, there is no single global guiding strategy for the management of NAFLD and NASH. The American Gastroenterological Association, in collaboration with 7 professional associations, convened an international conference comprising 32 experts in gastroenterology, hepatology, endocrinology, and primary care providers from the United States, Europe, Asia, and Australia. Conference content was informed by the results of a national NASH Needs Assessment Survey. The participants reviewed and discussed published literature on global burden, screening, risk stratification, diagnosis, and management of individuals with NAFLD, including those with NASH. Participants identified promising approaches for clinical practice and prepared a comprehensive, unified strategy for primary care providers and relevant specialists encompassing the full spectrum of NAFLD/NASH care. They also identified specific high-yield targets for clinical research and called for a unified, international public health response to NAFLD and NASH.

Predicting Non-Alcoholic Fatty Liver Disease for Adults Using Practical Clinical Measures: Evidence from the Multi-ethnic Study of Atherosclerosis

BACKGROUND

Many adults have risk factors for non-alcoholic fatty liver disease (NAFLD). Screening all adults with risk factors for NAFLD using imaging is not feasible.

OBJECTIVE

To develop a practical scoring tool for predicting NAFLD using participant demographics, medical history, anthropometrics, and lab values.

DESIGN

Cross-sectional.

PARTICIPANTS

Data came from 6194 white, African American, Hispanic, and Chinese American participants from the Multi-Ethnic Study of Atherosclerosis cohort, ages 45-85 years.

MAIN MEASURES

NAFLD was identified by liver computed tomography (≤ 40 Hounsfield units indicating > 30% hepatic steatosis) and data on 14 predictors was assessed for predicting NAFLD. Random forest variable importance was used to identify the minimum subset of variables required to achieve the highest predictive power. This subset was used to derive (n = 4132) and validate (n = 2063) a logistic regression-based score (NAFLD-MESA Index). A second NAFLD-Clinical Index excluding laboratory predictors was also developed.

KEY RESULTS

NAFLD prevalence was 6.2%. The model included eight predictors: age, sex, race/ethnicity, type 2 diabetes, smoking history, body mass index, gamma-glutamyltransferase (GGT), and triglycerides (TG). The NAFLD-Clinical Index model excluded GGT and TG. In the NAFLD-MESA model, the derivation set achieved an AUCNAFLD-MESA = 0.83 (95% CI, 0.81 to 0.86), and the validation set an AUCNAFLD-MESA = 0.80 (0.77 to 0.84). The NAFLD-Clinical Index model was AUCClinical = 0.78 [0.75 to 0.81] in the derivation set and AUCClinical = 0.76 [0.72 to 0.80] in the validation set (pBonferroni-adjusted < 0.01).

CONCLUSIONS

The two models are simple but highly predictive tools that can aid clinicians to identify individuals at high NAFLD risk who could benefit from imaging.

Preparing for the NASH Epidemic: A Call to Action

Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) are common conditions with a rising burden. Yet there are significant management gaps between clinical guidelines and practice in patients with NAFLD and NASH. Further, there is no single global guiding strategy for the management of NAFLD and NASH. The American Gastroenterological Association, in collaboration with 7 professional associations, convened an international conference comprising 32 experts in gastroenterology, hepatology, endocrinology, and primary care providers from the United States, Europe, Asia, and Australia. Conference content was informed by the results of a national NASH Needs Assessment Survey. The participants reviewed and discussed published literature on global burden, screening, risk stratification, diagnosis, and management of individuals with NAFLD, including those with NASH. Participants identified promising approaches for clinical practice and prepared a comprehensive, unified strategy for primary care providers and relevant specialists encompassing the full spectrum of NAFLD/NASH care. They also identified specific high-yield targets for clinical research and called for a unified, international public health response to NAFLD and NASH.

Preparing for the NASH epidemic: A call to action

Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) are common conditions with a rising burden. Yet there are significant management gaps between clinical guidelines and practice in patients with NAFLD and NASH. Further, there is no single global guiding strategy for the management of NAFLD and NASH. The American Gastroenterological Association, in collaboration with 7 professional associations, convened an international conference comprising 32 experts in gastroenterology, hepatology, endocrinology, and primary care providers from the United States, Europe, Asia, and Australia. Conference content was informed by the results of a national NASH Needs Assessment Survey. The participants reviewed and discussed published literature on global burden, screening, risk stratification, diagnosis, and management of individuals with NAFLD, including those with NASH. Participants identified promising approaches for clinical practice and prepared a comprehensive, unified strategy for primary care providers and relevant specialists encompassing the full spectrum of NAFLD/NASH care. They also identified specific high-yield targets for clinical research and called for a unified, international public health response to NAFLD and NASH.

Nonalcoholic fatty liver disease: use of diagnostic biomarkers and modalities in clinical practice

INTRODUCTION

The global burden of liver disease is increasing, and nonalcoholic fatty liver disease (NAFLD) is among the most common chronic liver diseases in Asia, Europe, North and South America. The field of noninvasive diagnostic and their role in staging, but also predicting outcome is evolving rapidly. There is a high-unmet need to stage patients with NAFLD and to identify the subset of patients at risk of progression to end-stage liver disease.

AREAS COVERED

The review covers all established diagnostic blood-based and imaging biomarkers to stage and grade NAFLD. Noninvasive surrogate scores are put into perspective of the available evidence and recommended use. The outlook includes genetics, combined algorithms, and artificial intelligence that will allow clinicians to guide and support the management in both early and later disease stages.

EXPERT OPINION

In the future, these diagnostics tests will help clinicians to establish patient care pathways and support the identification of relevant subgroups for monitoring and pharmacotherapy. In addition, researchers will be guided to better understand available scores and support the development of future prediction systems. These will likely include multiparametric aspects of the disease and machine learning algorithms will refine their use and integration with large datasets.

Current-Visit and Next-Visit Prediction for Fatty Liver Disease With a Large-Scale Dataset: Model Development and Performance Comparison

BACKGROUND

Fatty liver disease (FLD) arises from the accumulation of fat in the liver and may cause liver inflammation, which, if not well controlled, may develop into liver fibrosis, cirrhosis, or even hepatocellular carcinoma.

OBJECTIVE

We describe the construction of machine-learning models for current-visit prediction (CVP), which can help physicians obtain more information for accurate diagnosis, and next-visit prediction (NVP), which can help physicians provide potential high-risk patients with advice to effectively prevent FLD.

METHODS

The large-scale and high-dimensional dataset used in this study comes from Taipei MJ Health Research Foundation in Taiwan. We used one-pass ranking and sequential forward selection (SFS) for feature selection in FLD prediction. For CVP, we explored multiple models, including k-nearest-neighbor classifier (KNNC), Adaboost, support vector machine (SVM), logistic regression (LR), random forest (RF), Gaussian naïve Bayes (GNB), decision trees C4.5 (C4.5), and classification and regression trees (CART). For NVP, we used long short-term memory (LSTM) and several of its variants as sequence classifiers that use various input sets for prediction. Model performance was evaluated based on two criteria: the accuracy of the test set and the intersection over union/coverage between the features selected by one-pass ranking/SFS and by domain experts. The accuracy, precision, recall, F-measure, and area under the receiver operating characteristic curve were calculated for both CVP and NVP for males and females, respectively.

RESULTS

After data cleaning, the dataset included 34,856 and 31,394 unique visits respectively for males and females for the period 2009-2016. The test accuracy of CVP using KNNC, Adaboost, SVM, LR, RF, GNB, C4.5, and CART was respectively 84.28%, 83.84%, 82.22%, 82.21%, 76.03%, 75.78%, and 75.53%. The test accuracy of NVP using LSTM, bidirectional LSTM (biLSTM), Stack-LSTM, Stack-biLSTM, and Attention-LSTM was respectively 76.54%, 76.66%, 77.23%, 76.84%, and 77.31% for fixed-interval features, and was 79.29%, 79.12%, 79.32%, 79.29%, and 78.36%, respectively, for variable-interval features.

CONCLUSIONS

This study explored a large-scale FLD dataset with high dimensionality. We developed FLD prediction models for CVP and NVP. We also implemented efficient feature selection schemes for current- and next-visit prediction to compare the automatically selected features with expert-selected features. In particular, NVP emerged as more valuable from the viewpoint of preventive medicine. For NVP, we propose use of feature set 2 (with variable intervals), which is more compact and flexible. We have also tested several variants of LSTM in combination with two feature sets to identify the best match for male and female FLD prediction. More specifically, the best model for males was Stack-LSTM using feature set 2 (with 79.32% accuracy), whereas the best model for females was LSTM using feature set 1 (with 81.90% accuracy).

Current Options and Future Directions for NAFLD and NASH Treatment

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease worldwide, with a broad spectrum ranging from simple steatosis to advanced stage of nonalcoholic steatohepatitis (NASH). Although there are many undergoing clinical trials for NAFLD treatment, there is no currently approved treatment. NAFLD accounts as a major causing factor for the development of hepatocellular carcinoma (HCC), and its incidence rises accompanying the prevalence of obesity and diabetes. Reprogramming of antidiabetic and anti-obesity medicine is a major treatment option for NAFLD and NASH. Liver inflammation and cellular death, with or without fibrosis account for the progression of NAFLD to NASH. Therefore, molecules and signaling pathways involved in hepatic inflammation, fibrosis, and cell death are critically important targets for the therapy of NAFLD and NASH. In addition, the avoidance of aberrant infiltration of inflammatory cytokines by treating with CCR antagonists also provides a therapeutic option. Currently, there is an increasing number of pre-clinical and clinical trials undergoing to evaluate the effects of antidiabetic and anti-obesity drugs, antibiotics, pan-caspase inhibitors, CCR2/5 antagonists, and others on NAFLD, NASH, and liver fibrosis. Non-invasive serum diagnostic markers are developed for fulfilling the need of diagnostic testing in a large amount of NAFLD cases. Overall, a better understanding of the underlying mechanism of the pathogenesis of NAFLD is helpful to choose an optimized treatment.

A nomogram incorporated lifestyle indicators for predicting nonalcoholic fatty liver disease

Non-alcoholic fatty liver disease (NAFLD) is the most common chronic liver disease, and its pathogenesis is complicated and triggered by unbalanced diet, sedentary lifestyle, and genetic background. The aim of this study was to construct and validate a nomogram incorporated lifestyle habits for predicting NAFLD incidence.The overall cohort was divided into training set and test set as using computer-generated random numbers. We constructed the nomogram by multivariate logistic regression analysis in the training set. Thereafter, we validated this model by concordance index, the area under the receiver operating characteristic curve (ROC), net reclassification index, and a calibration curve in the test set. Additionally, we also evaluated the clinical usefulness of the nomogram by decision curve analysis.There were no statistically significant differences about characteristics between training cohort (n = 748) and test cohort (n = 320). Eleven features (age, sex, body mass index, drinking tea, physical exercise, energy, monounsaturated fatty acids, polyunsaturated fatty acids, hypertension, hyperlipidemia, diabetes) were incorporated to construct the nomogram, concordance index, the area under the ROC curve, net reclassification index were 0.801, 0.801, and 0.084, respectively, indicating the nomogram have good discrimination of predicting NAFLD incidence. Also, the calibration curve showed good consistency between nomogram prediction and actual probability. Moreover, the decision curve showed that when the threshold probability of an individual is within a range from approximately 0.5 to 0.8, this model provided more net benefit to predict NAFLD incidence risk than the current strategies.This nomogram can be regarded as a user-friendly tool for assessing the risk of NAFLD incidence, and thus help to facilitate management of NAFLD including lifestyle and medical interventions.

Colonoscopy and Risk of Colorectal Cancer in Patients With Nonalcoholic Fatty Liver Disease: A Retrospective Territory-Wide Cohort Study

The benefit of colonoscopy and/or polypectomy for colorectal cancer (CRC) prevention in patients with nonalcoholic fatty liver disease (NAFLD) remains unclear. We aimed to estimate the incidence rate of CRC in patients with NAFLD who had and had not undergone colonoscopy. We conducted a retrospective territory-wide cohort study for patients aged over 40 years with NAFLD identified with the International Classification of Diseases, Ninth Revision, Clinical Modification codes between January 1, 2000, and December 31, 2014. Patients were followed until CRC diagnosis, death, or December 31, 2017. We estimated CRC incidence and standardized incidence ratio (SIR) using the general population of Hong Kong as reference. We included 8,351 patients with NAFLD in the final analysis (median age, 56.2 years; interquartile ratio [IQR], 49.2-65.3 years; 45.4% male; median follow-up, 7.4 years; IQR, 5.4-9.6 years). Compared with the general population, patients with NAFLD who had not undergone colonoscopy had a higher incidence of CRC (SIR, 2.20; 95% confidence interval [CI], 1.64-2.88; P < 0.001). Patients with NAFLD who had undergone colonoscopy had a lower incidence of CRC (SIR, 0.54; 95% CI, 0.37-0.75; P < 0.001), especially among those aged above 50 years or with diabetes mellitus (DM). Patients with NAFLD with a high fibrosis-4 (FIB-4) score (>2.67) had a significantly higher risk of CRC after adjusting for demographic and metabolic factors. Conclusion: Patients with NAFLD who had undergone colonoscopy had a lower incidence of CRC than the general population, especially among those aged ≥50 years or with DM. A high FIB-4 index was associated with a higher risk of CRC.

Assessment of Classification Models and Relevant Features on Nonalcoholic Steatohepatitis Using Random Forest

Nonalcoholic fatty liver disease (NAFLD) is the hepatic manifestation of metabolic syndrome and is the most common cause of chronic liver disease in developed countries. Certain conditions, including mild inflammation biomarkers, dyslipidemia, and insulin resistance, can trigger a progression to nonalcoholic steatohepatitis (NASH), a condition characterized by inflammation and liver cell damage. We demonstrate the usefulness of machine learning with a case study to analyze the most important features in random forest (RF) models for predicting patients at risk of developing NASH. We collected data from patients who attended the Cardiovascular Risk Unit of Mostoles University Hospital (Madrid, Spain) from 2005 to 2021. We reviewed electronic health records to assess the presence of NASH, which was used as the outcome. We chose RF as the algorithm to develop six models using different pre-processing strategies. The performance metrics was evaluated to choose an optimized model. Finally, several interpretability techniques, such as feature importance, contribution of each feature to predictions, and partial dependence plots, were used to understand and explain the model to help obtain a better understanding of machine learning-based predictions. In total, 1525 patients met the inclusion criteria. The mean age was 57.3 years, and 507 patients had NASH (prevalence of 33.2%). Filter methods (the chi-square and Mann-Whitney-Wilcoxon tests) did not produce additional insight in terms of interactions, contributions, or relationships among variables and their outcomes. The random forest model correctly classified patients with NASH to an accuracy of 0.87 in the best model and to 0.79 in the worst one. Four features were the most relevant: insulin resistance, ferritin, serum levels of insulin, and triglycerides. The contribution of each feature was assessed via partial dependence plots. Random forest-based modeling demonstrated that machine learning can be used to improve interpretability, produce understanding of the modeled behavior, and demonstrate how far certain features can contribute to predictions.

Non-Alcoholic Fatty Liver Disease: Implementing Complete Automated Diagnosis and Staging. A Systematic Review

Background: Non-alcoholic fatty liver disease (NAFLD) is a fast-growing pathology around the world, being considered the most common chronic liver disease. It is diagnosed based on the presence of steatosis in more than 5% of hepatocytes without significant alcohol consumption. This review aims to provide a comprehensive overview of current studies of artificial intelligence (AI) applications that may help physicians in implementing a complete automated NAFLD diagnosis and staging. Methods: PubMed, EMBASE, Cochrane Library, and WILEY databases were screened for relevant publications in relation to AI applications in NAFLD. The search terms included: (non-alcoholic fatty liver disease OR NAFLD) AND (artificial intelligence OR machine learning OR neural networks OR deep learning OR automated diagnosis OR computer-aided diagnosis OR digital pathology OR automated ultrasound OR automated computer tomography OR automated magnetic imaging OR electronic health records). Results: Our search identified 37 articles about automated NAFLD diagnosis, out of which 15 articles analyzed imagistic techniques, 15 articles analyzed digital pathology, and 7 articles analyzed electronic health records (EHC). All studies included in this review show an accurate capacity of automated diagnosis and staging in NAFLD using AI-based software. Conclusions: We found significant evidence demonstrating that implementing a complete automated system for NAFLD diagnosis, staging, and risk stratification is currently possible, considering the accuracy, sensibility, and specificity of available AI-based tools.

Application of Artificial Intelligence for the Diagnosis and Treatment of Liver Diseases

Modern medical care produces large volumes of multimodal patient data, which many clinicians struggle to process and synthesize into actionable knowledge. In recent years, artificial intelligence (AI) has emerged as an effective tool in this regard. The field of hepatology is no exception, with a growing number of studies published that apply AI techniques to the diagnosis and treatment of liver diseases. These have included machine-learning algorithms (such as regression models, Bayesian networks, and support vector machines) to predict disease progression, the presence of complications, and mortality; deep-learning algorithms to enable rapid, automated interpretation of radiologic and pathologic images; and natural-language processing to extract clinically meaningful concepts from vast quantities of unstructured data in electronic health records. This review article will provide a comprehensive overview of hepatology-focused AI research, discuss some of the barriers to clinical implementation and adoption, and suggest future directions for the field.

Artificial intelligence/neural network system for the screening of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis

AIM

We aimed to develop a novel noninvasive test using an artificial intelligence (AI)/neural network (NN) system (named nonalcoholic steatohepatitis [NASH]-Scope) to screen nonalcoholic fatty liver disease (NAFLD) and NASH.

METHODS

We enrolled 324 and 74 patients histologically diagnosed with NAFLD for training and validation studies, respectively. Two independent pathologists histologically diagnosed patients with NAFLD for validation study. Additionally, 48 subjects who underwent a medical health checkup and did not show fatty liver ultrasonographically and had normal serum aminotransferase levels were categorized as the non-NAFLD group. NASH-Scope was based on 11 clinical values: age, sex, height, weight, waist circumference, aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transferase, cholesterol, triglyceride, and platelet count.

RESULTS

The sensitivity, specificity, positive predictive value, negative predictive value, and area under the receiver operator characteristic curve of NASH-Scope for distinguishing NAFLD from non-NAFLD in the training study and validation study were 99.7% versus 97.2%, 97.8% versus 97.8%, 99.7% versus 98.6%, 97.8% versus 95.7%, and 0.999 versus 0.950, respectively. Those for distinguishing NASH with fibrosis from NAFLD without fibrosis were 99.5% versus 90.7%, 84.3% versus 93.3%, 94.2% versus 98.0%, 98.6% versus 73.7%, and 0.960 versus 0.950. These results were excellent, even when the output data were divided into two categories without any gray zone.

CONCLUSIONS

The AI/NN system, termed as NASH-Scope, is practical and can accurately differentially diagnose between NAFLD and non-NAFLD and between NAFLD without fibrosis and NASH with fibrosis. Thus, NASH-Scope is useful for screening nonalcoholic fatty liver and NASH.

Fatty Liver Disease Prediction Model Based on Big Data of Electronic Physical Examination Records

Fatty liver disease (FLD) is a common liver disease, which poses a great threat to people's health, but there is still no optimal method that can be used on a large-scale screening. This research is based on machine learning algorithms, using electronic physical examination records in the health database as data support, to a predictive model for FLD. The model has shown good predictive ability on the test set, with its AUC reaching 0.89. Since there are a large number of electronic physical examination records in most of health database, this model might be used as a non-invasive diagnostic tool for FLD for large-scale screening.

Recent Applications of Artificial Intelligence in detection of Gastrointestinal, Hepatic and Pancreatic Diseases

There is substantial progress in artificial intelligence (AI) algorithms and their medical sciences applications in the last two decades. AI-assisted programs have already been established for remotely health monitoring using sensors and smartphones. A variety of AI-based prediction models available for the gastrointestinal inflammatory, non-malignant diseases, and bowel bleeding using wireless capsule endoscopy, electronic medical records for hepatitis-associated fibrosis, pancreatic carcinoma using endoscopic ultrasounds. AI-based models may be of immense help for healthcare professionals in the identification, analysis, and decision support using endoscopic images to establish prognosis and risk assessment of patient's treatment using multiple factors. Although enough randomized clinical trials are warranted to establish the efficacy of AI-algorithms assisted and non-AI based treatments before approval of such techniques from medical regulatory authorities. In this article, available AI approaches and AI-based prediction models for detecting gastrointestinal, hepatic, and pancreatic diseases are reviewed. The limitation of AI techniques in such disease prognosis, risk assessment, and decision support are discussed.

High-dimensional hepatopath data analysis by machine learning for predicting HBV-related fibrosis

Chronic HBV infection, the main cause of liver cirrhosis and hepatocellular carcinoma, has become a global health concern. Machine learning algorithms are particularly adept at analyzing medical phenomenon by capturing complex and nonlinear relationships in clinical data. Our study proposed a predictive model on the basis of 55 routine laboratory and clinical parameters by machine learning algorithms as a novel non-invasive method for liver fibrosis diagnosis. The model was further evaluated on the accuracy and rationality and proved to be highly accurate and efficient for the prediction of HBV-related fibrosis. In conclusion, we suggested a potential combination of high-dimensional clinical data and machine learning predictive algorithms for the liver fibrosis diagnosis.

Artificial intelligence in precision medicine in hepatology

The advancement of investigation tools and electronic health records (EHR) enables a paradigm shift from guideline-specific therapy toward patient-specific precision medicine. The multiparametric and large detailed information necessitates novel analyses to explore the insight of diseases and to aid the diagnosis, monitoring, and outcome prediction. Artificial intelligence (AI), machine learning, and deep learning (DL) provide various models of supervised, or unsupervised algorithms, and sophisticated neural networks to generate predictive models more precisely than conventional ones. The data, application tasks, and algorithms are three key components in AI. Various data formats are available in daily clinical practice of hepatology, including radiological imaging, EHR, liver pathology, data from wearable devices, and multi-omics measurements. The images of abdominal ultrasonography, computed tomography, and magnetic resonance imaging can be used to predict liver fibrosis, cirrhosis, non-alcoholic fatty liver disease (NAFLD), and differentiation of benign tumors from hepatocellular carcinoma (HCC). Using EHR, the AI algorithms help predict the diagnosis and outcomes of liver cirrhosis, HCC, NAFLD, portal hypertension, varices, liver transplantation, and acute liver failure. AI helps to predict severity and patterns of fibrosis, steatosis, activity of NAFLD, and survival of HCC by using pathological data. Despite of these high potentials of AI application, data preparation, collection, quality, labeling, and sampling biases of data are major concerns. The selection, evaluation, and validation of algorithms, as well as real-world application of these AI models, are also challenging. Nevertheless, AI opens the new era of precision medicine in hepatology, which will change our future practice.

Artificial intelligence in prediction of non-alcoholic fatty liver disease and fibrosis

Artificial intelligence (AI) has become increasingly widespread in our daily lives, including healthcare applications. AI has brought many new insights into better ways we care for our patients with chronic liver disease, including non-alcoholic fatty liver disease and liver fibrosis. There are multiple ways to apply the AI technology on top of the conventional invasive (liver biopsy) and noninvasive (transient elastography, serum biomarkers, or clinical prediction models) approaches. In this review article, we discuss the principles of applying AI on electronic health records, liver biopsy, and liver images. A few common AI approaches include logistic regression, decision tree, random forest, and XGBoost for data at a single time stamp, recurrent neural networks for sequential data, and deep neural networks for histology and images.

From heterogeneous healthcare data to disease-specific biomarker networks: A hierarchical Bayesian network approach

In this work, we introduce an entirely data-driven and automated approach to reveal disease-associated biomarker and risk factor networks from heterogeneous and high-dimensional healthcare data. Our workflow is based on Bayesian networks, which are a popular tool for analyzing the interplay of biomarkers. Usually, data require extensive manual preprocessing and dimension reduction to allow for effective learning of Bayesian networks. For heterogeneous data, this preprocessing is hard to automatize and typically requires domain-specific prior knowledge. We here combine Bayesian network learning with hierarchical variable clustering in order to detect groups of similar features and learn interactions between them entirely automated. We present an optimization algorithm for the adaptive refinement of such group Bayesian networks to account for a specific target variable, like a disease. The combination of Bayesian networks, clustering, and refinement yields low-dimensional but disease-specific interaction networks. These networks provide easily interpretable, yet accurate models of biomarker interdependencies. We test our method extensively on simulated data, as well as on data from the Study of Health in Pomerania (SHIP-TREND), and demonstrate its effectiveness using non-alcoholic fatty liver disease and hypertension as examples. We show that the group network models outperform available biomarker scores, while at the same time, they provide an easily interpretable interaction network.

High-dimensional and heterogeneous healthcare data, such as electronic health records or epidemiological study data, contain much information on yet unknown risk factors that are associated with disease development. The identification of these risk factors may help to improve prevention, diagnosis, and therapy. Bayesian networks are powerful statistical models that can decipher these complex relationships. However, high dimensionality and heterogeneity of data, together with missing values and high feature correlation, make it difficult to automatically learn a good model from data. To facilitate the use of network models, we present a novel, fully automated workflow that combines network learning with hierarchical clustering. The algorithm reveals groups of strongly related features and models the interactions among those groups. It results in simpler network models that are easier to analyze. We introduce a method of adaptive refinement of such models to ensure that disease-relevant parts of the network are modeled in great detail. Our approach makes it easy to learn compact, accurate, and easily interpretable biomarker interaction networks. We test our method extensively on simulated data as well as data from the Study of Health in Pomerania (SHIP-Trend) by learning models of hypertension and non-alcoholic fatty liver disease.

Performance Assessment of Classification Algorithms on Early Detection of Liver Syndrome

In the recent era, a liver syndrome that causes any damage in life capacity is exceptionally normal everywhere throughout the world. It has been found that liver disease is exposed more in young people as a comparison with other aged people. At the point when liver capacity ends up, life endures just up to 1 or 2 days scarcely, and it is very hard to predict such illness in the early stage. Researchers are trying to project a model for early prediction of liver disease utilizing various machine learning approaches. However, this study compares ten classifiers including A1DE, NB, MLP, SVM, KNN, CHIRP, CDT, Forest-PA, J48, and RF to find the optimal solution for early and accurate prediction of liver disease. The datasets utilized in this study are taken from the UCI ML repository and the GitHub repository. The outcomes are assessed via RMSE, RRSE, recall, specificity, precision, G-measure, F-measure, MCC, and accuracy. The exploratory outcomes show a better consequence of RF utilizing the UCI dataset. Assessing RF using RMSE and RRSE, the outcomes are 0.4328 and 87.6766, while the accuracy of RF is 72.1739% that is also better than other employed classifiers. However, utilizing the GitHub dataset, SVM beats other employed techniques in terms of increasing accuracy up to 71.3551%. Moreover, the comprehensive outcomes of this exploration can be utilized as a reference point for further research studies that slight assertion concerning the enhancement in extrapolation through any new technique, model, or framework can be benchmarked and confirmed.

Non-alcoholic Fatty Liver and Liver Fibrosis Predictive Analytics: Risk Prediction and Machine Learning Techniques for Improved Preventive Medicine

Non-alcoholic fatty liver disease (NAFLD) is the most common liver disease worldwide, with a prevalence of 20%-30% in the general population. NAFLD is associated with increased risk of cardiovascular disease and may progress to cirrhosis with time. The purpose of this study was to predict the risks associated with NAFLD and advanced fibrosis on the Fatty Liver Index (FLI) and the 'NAFLD fibrosis 4' calculator (FIB-4), to enable physicians to make more optimal preventive medical decisions. A prospective cohort of apparently healthy volunteers from the Tel Aviv Medical Center Inflammation Survey (TAMCIS), admitted for their routine annual health check-up. Data from the TAMCIS database were subjected to machine learning classification models to predict individual risk after extensive data preparation that included the computation of independent variables over several time points. After incorporating the time covariates and other key variables, this technique outperformed the predictive power of current popular methods (an improvement in AUC above 0.82). New powerful factors were identified during the predictive process. The findings can be used for risk stratification and in planning future preventive strategies based on lifestyle modifications and medical treatment to reduce the disease burden. Interventions to prevent chronic disease can substantially reduce medical complications and the costs of the disease. The findings highlight the value of predictive analytic tools in health care environments. NAFLD constitutes a growing burden on the health system; thus, identification of the factors related to its incidence can make a strong contribution to preventive medicine.

Application of Machine Learning for Tumor Growth Inhibition - Overall Survival Modeling Platform

Machine learning (ML) was used to leverage tumor growth inhibition (TGI) metrics to characterize the relationship with overall survival (OS) as a novel approach and to compare with traditional TGI‐OS modeling methods. Historical dataset from a phase III non‐small cell lung cancer study (OAK, atezolizumab vs. docetaxel, N = 668) was used. ML methods support the validity of TGI metrics in predicting OS. With lasso, the best model with TGI metrics outperforms the best model without TGI metrics. Boosting was the best linear ML method for this dataset with reduced estimation bias and lowest Brier score, suggesting better prediction accuracy. Random forest did not outperform linear ML methods despite hyperparameter optimization. Kernel machine was marginally the best nonlinear ML method for this dataset and uncovered nonlinear and interaction effects. Nonlinear ML may improve prediction by capturing nonlinear effects and covariate interactions, but its predictive performance and value need further evaluation with larger datasets.

A Nomogram Model Based on Noninvasive Bioindicators to Predict 3-Year Risk of Nonalcoholic Fatty Liver in Nonobese Mainland Chinese: A Prospective Cohort Study

The purpose of this study is to establish and validate an accurate and personalized nonalcoholic fatty liver disease (NAFLD) prediction model based on the nonobese population in China. This study is a secondary analysis of a prospective study. We included 6,155 nonobese adults without NAFLD at baseline, with a median follow-up of 2.3 years. Univariate and multivariate Cox regression analyses were used to determine independent predictors. The least absolute shrinkage and selection operator (LASSO) regression analysis was used to optimize the selection of variables. Based on the results of multivariate analysis, a prediction model was established. Harrell's consistency index (C-index) and area under the curve (AUC) were used to determine the discrimination of the proposed model. The goodness of fit of the calibration model was tested, and the clinical application value of the model was evaluated by decision curve analysis (DCA). The participants were randomly divided into a training cohort (n = 4,605) and a validation cohort (n = 1,550). Finally, seven of the variables (HDL-c, BMI, GGT, ALT, TB, DBIL, and TG) were included in the prediction model. In the training cohort, the C-index and AUC value of this prediction model were 0.832 (95% confidence interval (CI), 0.820-0.844) and 0.861 (95% CI, 0.849-0.873), respectively. In the validation cohort, the C-index and AUC values of this prediction model were 0.829 (95% CI, 0.806-0.852) and 0.859 (95% CI, 0.841-0.877), respectively. The calibration plots demonstrated good agreement between the estimated probability and the actual observation. DCA demonstrated a clinically effective predictive model. Our nomogram can be used as a simple, reasonable, economical, and widely used tool to predict the 3-year risk of NAFLD in nonobese populations in China, which is helpful for timely intervention and reducing the incidence of NAFLD.

Comparative effectiveness of glucagon-like peptide-1 receptor agonists versus dipeptidyl peptidase-4 inhibitors on noninvasive indices of hepatic steatosis and fibrosis in patients with type 2 diabetes mellitus

BACKGROUND AND AIMS

Nonalcoholic fatty liver disease (NAFLD) is highly prevalent in patients with type 2 diabetes mellitus (T2DM). There is currently no approved treatment for NAFLD. The main aim was the evaluation of the effect of glucagon-like peptide-1 receptor agonists (GLP-1 RA) vs. dipeptidyl peptidase-4 inhibitor (DPP-4i) treatment on noninvasive indices of hepatic steatosis and fibrosis in patients with T2DM.

METHODS

In this retrospective study, three noninvasive indices of hepatic steatosis [HSI, NAFLD ridge score, and triglycerides (TG) to high-density lipoprotein cholesterol (HDL-C) ratio] and five of fibrosis (APRI, FIB-4, NAFLD fibrosis score, BAAT and BARD) were calculated before and after (6-18 months) the addition of a DPP-4i (n = 152) or a GLP-1 RA (n = 37) in patients with T2DM.

RESULTS

Regarding steatosis indices, NAFLD ridge score was significantly decreased in the GLP-1 RA group (baseline: 0.90 ± 0.34, follow-up: 0.67 ± 0.24; p = 0.001), but not in the DPP-4i group (p = 0.25); the difference for group∗time interaction was significant (p = 0.02). HSI showed a trend between groups, being significantly different at baseline and follow-up (p < 0.001) with no significant difference in group∗time interaction. Indices of fibrosis were not essentially changed within or between groups.

CONCLUSIONS

NAFLD ridge score was significantly decreased after the addition of GLP-1 RA in patients with T2DM. This study warrants further prospective clinical trials.

Selecting the best machine learning algorithm to support the diagnosis of Non-Alcoholic Fatty Liver Disease: A meta learner study

Background & aims

Liver ultrasound scan (US) use in diagnosing Non-Alcoholic Fatty Liver Disease (NAFLD) causes costs and waiting lists overloads. We aimed to compare various Machine learning algorithms with a Meta learner approach to find the best of these as a predictor of NAFLD.

Methods

The study included 2970 subjects, 2920 constituting the training set and 50, randomly selected, used in the test phase, performing cross-validation. The best predictors were combined to create three models: 1) FLI plus GLUCOSE plus SEX plus AGE, 2) AVI plus GLUCOSE plus GGT plus SEX plus AGE, 3) BRI plus GLUCOSE plus GGT plus SEX plus AGE. Eight machine learning algorithms were trained with the predictors of each of the three models created. For these algorithms, the percent accuracy, variance and percent weight were compared.

Results

The SVM algorithm performed better with all models. Model 1 had 68% accuracy, with 1% variance and an algorithm weight of 27.35; Model 2 had 68% accuracy, with 1% variance and an algorithm weight of 33.62 and Model 3 had 77% accuracy, with 1% variance and an algorithm weight of 34.70. Model 2 was the most performing, composed of AVI plus GLUCOSE plus GGT plus SEX plus AGE, despite a lower percentage of accuracy.

Conclusion

A Machine Learning approach can support NAFLD diagnosis and reduce health costs. The SVM algorithm is easy to apply and the necessary parameters are easily retrieved in databases.

A New Endemic of Concomitant Nonalcoholic Fatty Liver Disease and Chronic Hepatitis B

Hepatitis B virus (HBV) infection remains a global public problem despite the availability of an effective vaccine. In the past decades, nonalcoholic fatty liver disease (NAFLD) has surpassed HBV as the most common cause of chronic liver disease worldwide. The prevalence of concomitant chronic hepatitis B (CHB) and NAFLD thus reaches endemic proportions in geographic regions where both conditions are common. Patients with CHB and NAFLD are at increased risk of liver disease progression to cirrhosis and hepatocellular carcinoma. Due to the complexity of the pathogenesis, accurate diagnosis of NAFLD in CHB patients can be challenging. Liver biopsy is considered the gold standard for diagnosing and determining disease severity, but it is an invasive procedure with potential complications. There is a growing body of literature on the application of novel noninvasive serum biomarkers and advanced radiological modalities to diagnose and evaluate NAFLD, but most have not been adequately validated, especially for patients with CHB. Currently, there is no approved therapy for NAFLD, although many new agents are in different phases of development. This review provides a summary of the epidemiology, clinical features, diagnosis, and management of the NAFLD and highlights the unmet needs in the areas of CHB and NAFLD coexistence.

Diagnostic and interventional circulating biomarkers in nonalcoholic steatohepatitis

INTRODUCTION

In the setting of the obesity epidemic, nonalcoholic fatty liver disease (NAFLD) has become one of the most prevalent forms of chronic liver disease worldwide. Approximately 25% of adults globally have NAFLD which includes those with NAFL, or simple steatosis, and individuals with nonalcoholic steatohepatitis (NASH) where inflammation, hepatocyte injury and potentially hepatic fibrosis are found in conjunction with steatosis. Individuals with NASH, particularly those with hepatic fibrosis, have higher rates of liver-related and overall mortality, making this distinction of significant clinical importance. One of the core challenges in current clinical practice is identifying this subset of individuals with NASH without the use of liver biopsy, the gold standard for both diagnostics and staging disease severity. Identifying noninvasive biomarkers, an accurately measured and reproducible parameter, would aide in identifying patients eligible for NASH pharmacotherapy clinical trials and to help tailor intensity of monitoring required.

METHODS RESULTS AND CONCLUSIONS

In this review, we highlight both the currently available and novel diagnostic and interventional circulating biomarkers under investigation for NASH, underscoring their accuracy and limitations relevant to our patient population and current clinical practice.

Development and Validation of a Simple Index Based on Non-Enhanced CT and Clinical Factors for Prediction of Non-Alcoholic Fatty Liver Disease

OBJECTIVE

A widely applicable, non-invasive screening method for non-alcoholic fatty liver disease (NAFLD) is needed. We aimed to develop and validate an index combining computed tomography (CT) and routine clinical data for screening for NAFLD in a large cohort of adults with pathologically proven NAFLD.

MATERIALS AND METHODS

This retrospective study included 2218 living liver donors who had undergone liver biopsy and CT within a span of 3 days. Donors were randomized 2:1 into development and test cohorts. CTL-S was measured by subtracting splenic attenuation from hepatic attenuation on non-enhanced CT. Multivariable logistic regression analysis of the development cohort was utilized to develop a clinical-CT index predicting pathologically proven NAFLD. The diagnostic performance was evaluated by analyzing the areas under the receiver operating characteristic curve (AUC). The cutoffs for the clinical-CT index were determined for 90% sensitivity and 90% specificity in the development cohort, and their diagnostic performance was evaluated in the test cohort.

RESULTS

The clinical-CT index included CT_L-S, body mass index, and aspartate transaminase and triglyceride concentrations. In the test cohort, the clinical-CT index (AUC, 0.81) outperformed CTL-S (0.74; p < 0.001) and clinical indices (0.73-0.75; p < 0.001) in diagnosing NAFLD. A cutoff of ≥ 46 had a sensitivity of 89% and a specificity of 41%, whereas a cutoff of ≥ 56.5 had a sensitivity of 57% and a specificity of 89%.

CONCLUSION

The clinical-CT index is more accurate than CTL-S and clinical indices alone for the diagnosis of NAFLD and may be clinically useful in screening for NAFLD.

Pattern and impact of hepatic adverse events encountered during immune checkpoint inhibitors - A territory-wide cohort study

Background

Immune checkpoint inhibitors (ICIs) are increasingly used in the treatment of cancers. We aimed to evaluate the incidence and prognostic impact of hepatic adverse events (AEs) in a territory‐wide cohort of patients who received ICIs.

Methods

Patients were identified from a territory‐wide database who received ICIs in 2014‐2018. Hepatic AEs were defined as any elevation of liver biochemistries including serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), or total bilirubin levels. Hepatic AEs were graded according to Common Terminology Criteria for Adverse Events (CTCAE) v5.0.

Results

Total of 1480 patients were identified (mean age 60 years, male 65.5%) and the commonest malignancies being lung cancer (39.6%), liver cancer (16.5%), and gastrointestinal cancer (10.0%). Grade 1‐2 and grade 3‐4 hepatic AEs occurred in 41.3% and 14.9% of patients during ICI treatment, respectively. Patients with liver cancer had the highest rate of hepatic AEs (grade 1‐2:54.1%; grade 3‐4:32.8%). Among 711 patients with hepatic AEs, 383 (53.9%) had raised ALT/AST only, and 328 (46.1%) had concomitant raised ALT/AST and bilirubin levels. In the whole cohort, median overall survival of patients without any hepatic AEs, grade 1‐2 and grade 3‐4 hepatic AEs during ICI treatment was 9.0 months, 7.2 months, and 3.3 months (P < .001), respectively. Similar results on overall survival were obtained among different types of cancers.

Conclusions

Hepatic AEs occur in more than half of patients receiving ICIs for cancer treatment, with approximately 15% being grade 3‐4 AEs. Occurrence of hepatic AEs is associated with worse prognosis.

Risk Prediction for Non-alcoholic Fatty Liver Disease Based on Biochemical and Dietary Variables in a Chinese Han Population

Nonalcoholic fatty liver disease (NAFLD) is a common liver disease globally, but there are no optimal methods for its prediction or diagnosis. The present cross-sectional study proposes a non-invasive tool for NAFLD screening. The study included 2,446 individuals, of whom 574 were NAFLD patients. Multivariable logistic regression analysis was used to identify risk factors for NAFLD and incorporate them in a risk prediction nomogram model; the variables included both clinical and lifestyle-related variables. Following stepwise regression, BMI, waist circumference, serum triglyceride, high-density lipoprotein cholesterol, alanine aminotransferase, presence of diabetes and hyperuricemia, tuber and fried food consumption were identified as significant risk factors and used in the model. The final nomogram was found to have good discrimination ability (area under the receiver operating characteristic curve = 0.843 [95% CI: 0.819-0.867]), and reasonable accuracy for the prediction of NAFLD risk. A cut-off score of <180 for the nomogram was found to have high sensitivity and predictivity for the exclusion of individuals from screening. The model can be used as a non-invasive tool for mass screening.

Supervised Machine Learning in Oncology: A Clinician's Guide

The widespread adoption of electronic health records has resulted in an abundance of imaging and clinical information. New data-processing technologies have the potential to revolutionize the practice of medicine by deriving clinically meaningful insights from large-volume data. Among those techniques is supervised machine learning, the study of computer algorithms that use self-improving models that learn from labeled data to solve problems. One clinical area of application for supervised machine learning is within oncology, where machine learning has been used for cancer diagnosis, staging, and prognostication. This review describes a framework to aid clinicians in understanding and critically evaluating studies applying supervised machine learning methods. Additionally, we describe current studies applying supervised machine learning techniques to the diagnosis, prognostication, and treatment of cancer, with a focus on gastroenterological cancers and other related pathologies.

Applying Machine Learning in Liver Disease and Transplantation: A Comprehensive Review

Machine learning (ML) utilizes artificial intelligence to generate predictive models efficiently and more effectively than conventional methods through detection of hidden patterns within large data sets. With this in mind, there are several areas within hepatology where these methods can be applied. In this review, we examine the literature pertaining to machine learning in hepatology and liver transplant medicine. We provide an overview of the strengths and limitations of ML tools and their potential applications to both clinical and molecular data in hepatology. ML has been applied to various types of data in liver disease research, including clinical, demographic, molecular, radiological, and pathological data. We anticipate that use of ML tools to generate predictive algorithms will change the face of clinical practice in hepatology and transplantation. This review will provide readers with the opportunity to learn about the ML tools available and potential applications to questions of interest in hepatology.

Screening for non-alcoholic fatty liver disease in type 2 diabetes using non-invasive scores and association with diabetic complications

OBJECTIVE

Non-alcoholic fatty liver disease (NAFLD) is prevalent in patients with type 2 diabetes. Here, we estimate the proportion of patients with type 2 diabetes that should be referred to hepatologists according to the European Association for the Study of the Liver (EASL)-European Association for the Study of Diabetes (EASD)-European Association for the Study of Obesity (EASO) Guidelines and evaluate the association between non-invasive biomarkers of steatosis and fibrosis and diabetic complications.

RESEARCH DESIGN AND METHODS

This is a retrospective analysis of type 2 diabetes patients who attended on a regular basis our diabetes clinic between 2013 and 2018 (n=2770). Steatosis was assessed using Fatty Liver Index (FLI), Hepatic Steatosis Index and NAFLD Ridge Score and fibrosis using NAFLD Fibrosis Score (NFS), Fibrosis-4 (FIB-4), aspartate aminotransferase (AST) to platelet ratio index (APRI) and AST/alanine aminotransferase (ALT) ratio. Outcome measures were altered albumin excretion rate (AER), chronic kidney disease (CKD) and cardiovascular disease (CVD).

RESULTS

The prevalence of advanced fibrosis varied from 1% (APRI) to 33% (NFS). The application of the guidelines using a sequential combination of FLI and FIB-4 would lead to referral of 28.3% of patients when using standard FIB-4 cut-offs, while this number dropped to 13.4% when age-adjusted FIB-4 thresholds were applied. A higher prevalence of altered AER was associated with liver steatosis (FLI: OR: 3.49; 95% CI 2.05 to 5.94, p<0.01), whereas liver fibrosis was associated with CKD (FIB-4: OR: 6.39; 95% CI 4.05 to 10.08, p<0.01) and CVD (FIB-4: OR: 2.62; 95% CI 1.69 to 4.04, p<0.01).

CONCLUSIONS

While specific fibrosis scores identify different proportion of patients with advanced fibrosis, the use of age-adjusted FIB-4 cut-offs leads to a drop in gray-zone results, making referrals to hepatologists more sustainable. Interestingly non-invasive biomarkers were consistently associated with a different pattern of diabetic complications.

Risk of hepatitis B surface antigen seroreversion after corticosteroid treatment in patients with previous hepatitis B virus exposure

BACKGROUND & AIMS

Systemic corticosteroids may cause HBV reactivation, but the impact on patients with previous HBV exposure is poorly defined. We aimed to study the risk of HBsAg seroreversion and hepatitis flare in patients with previous HBV exposure.

METHODS

Patients who were negative for HBsAg and received corticosteroids between 2001-2010 were included. Patients who were positive for antibody to HBsAg (anti-HBs) and/or to HBcAg (anti-HBc) were defined as having previous HBV exposure. The primary endpoint was HBsAg seroreversion; the secondary endpoint was hepatitis flare (alanine aminotransferase >80 U/L) at 1 year.

RESULTS

A total of 12,997 patients fulfilled the inclusion criteria: anti-HBs positive only (n = 10,561); anti-HBc positive only (n = 970); anti-HBs & anti-HBc positive (n = 830) and anti-HBs & anti-HBc negative (n = 636). HBsAg seroreversion occurred in 165 patients. Patients who were anti-HBc positive only had a higher risk of HBsAg seroreversion (1-year incidence 1.8%) than those negative for both anti-HBs & anti-HBc (0%; p = 0.014). Patients with previous HBV exposure had a similarly low risk of liver failure as unexposed individuals (1.1% vs. 0.9%). The risk of a hepatitis flare started to increase in those receiving corticosteroids at peak daily doses of 20-40 mg (adjusted hazard ratio [HR] 2.19, p = 0.048) or >40 mg (aHR 2.11, p = 0.015) prednisolone equivalents for <7 days, and was increased at treatment durations of 7-28 days and >28 days (aHR 2.02-3.85; p <0.001-0.012).

CONCLUSIONS

In HBsAg-negative patients who were only anti-HBc positive, high peak daily doses of corticosteroids increased the risk of hepatitis flare, but not seroreversion. The rate of liver failure was low and similar in HBV exposed and unexposed individuals; there were no deaths, nor any requirement for liver transplantation.

LAY SUMMARY

It is important to know the hepatitis B virus (HBV) status before starting corticosteroid therapy. Patients with resolved HBV infection without detectable immunity are at an increased risk of HBV surface antigen seroreversion after corticosteroid therapy. High peak daily doses of corticosteroids (>40 mg prednisolone equivalents) increase the risk of hepatitis flare, but not seroreversion, in patients with previous exposure to HBV, irrespective of the duration of treatment. Interval monitoring of liver biochemistries is essential for the early detection of hepatitis flares in these patients.

Confounding factors of non-invasive tests for nonalcoholic fatty liver disease

Nonalcoholic fatty liver disease (NAFLD) affects at least 25% of the general adult population worldwide. Because only a fraction of the patients would develop liver-related complications, it is preferable to perform non-invasive tests as the initial assessment. This review summarizes the known and potential confounding factors that affect the performance of non-invasive tests of hepatic steatosis and fibrosis in patients with NAFLD. Clinicians may apply the knowledge and exercise caution in selecting investigations and interpreting test results when confounding factors are present.

Application of Artificial Intelligence to Gastroenterology and Hepatology

Since 2010, substantial progress has been made in artificial intelligence (AI) and its application to medicine. AI is explored in gastroenterology for endoscopic analysis of lesions, in detection of cancer, and to facilitate the analysis of inflammatory lesions or gastrointestinal bleeding during wireless capsule endoscopy. AI is also tested to assess liver fibrosis and to differentiate patients with pancreatic cancer from those with pancreatitis. AI might also be used to establish prognoses of patients or predict their response to treatments, based on multiple factors. We review the ways in which AI may help physicians make a diagnosis or establish a prognosis and discuss its limitations, knowing that further randomized controlled studies will be required before the approval of AI techniques by the health authorities.

Homeostatic model assessment of insulin resistance closely related to lobular inflammation in nonalcoholic fatty liver disease

OBJECTIVE

Insulin resistance (IR) has been established as a major risk factor for nonalcoholic fatty liver disease (NAFLD) where it exerts effects on plasma glucose homeostasis, cellular anabolism, and organ glucose uptake. Owing to paucity of studies focused on peripheral IR in relation to pathological outcome, we aim to investigate homeostatic model assessment of insulin resistance (HOMA-IR) by histological characteristics of NAFLD.

METHODS

Liver biopsy of 588 patients was screened. After excluding etiologies other than NAFLD and factors contributing to IR, serum HOMA-IR was compared with patients' histologic features. Univariate and multivariate analyses were conducted to assess their relationship. Area under the receiver operating characteristic (AUROC) was calculated to assess the discriminatory ability of homeostatic model assessment of IR for advanced lobular inflammation (LI).

RESULTS

We observed higher serum level of alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transferase, and low-density lipoprotein as HOMA-IR increased. HOMA-IR is significantly associated with severity of LI (odds ratio = 1.222, 95% confidence interval = 1.135-1.315, P < 0.001), similar association remained after adjusting for age, BMI, hemoglobin A1c, alanine aminotransferase, aspartate aminotransferase, low-density lipoprotein, high-density lipoprotein, and triglycerides (odds ratio = 1.205, 95% confidence interval = 1.102-1.317, P < 0.001). HOMA-IR is discriminant of LI with AUROC = 0.832 and cutoff = 2.995 (sensitivity = 0.938, specificity = 0.569).

CONCLUSION

This study demonstrated a strong and independent association of HOMA-IR with the severity of liver inflammation by histological evaluation in NAFLD patients without diabetes or metabolic syndrome, and its possible role in diagnosis of LI could be translated into clinical assessment of NAFLD patients with uncertainty of nonalcoholic steatohepatitis progression.

Current Status in Testing for Nonalcoholic Fatty Liver Disease (NAFLD) and Nonalcoholic Steatohepatitis (NASH)

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in Western countries with almost 25% affected adults worldwide. The growing public health burden is getting evident when considering that NAFLD-related liver transplantations are predicted to almost double within the next 20 years. Typically, hepatic alterations start with simple steatosis, which easily progresses to more advanced stages such as nonalcoholic steatohepatitis (NASH), fibrosis and cirrhosis. This course of disease finally leads to end-stage liver disease such as hepatocellular carcinoma, which is associated with increased morbidity and mortality. Although clinical trials show promising results, there is actually no pharmacological agent approved to treat NASH. Another important problem associated with NASH is that presently the liver biopsy is still the gold standard in diagnosis and for disease staging and grading. Because of its invasiveness, this technique is not well accepted by patients and the method is prone to sampling error. Therefore, an urgent need exists to find reliable, accurate and noninvasive biomarkers discriminating between different disease stages or to develop innovative imaging techniques to quantify steatosis.