Detecting liver fibrosis using a machine learning-based approach to the quantification of the heart-induced deformation in tagged MR images

Artificial Intelligence in Liver Diseases: Recent Advances

Liver diseases cause a significant burden on public health worldwide. In spite of great advances during recent years, there are still many challenges in the diagnosis and treatment of liver diseases. During recent years, artificial intelligence (AI) has been widely used for the diagnosis, risk stratification, and prognostic prediction of various diseases based on clinical datasets and medical images. Accumulative studies have shown its performance for diagnosing patients with nonalcoholic fatty liver disease and liver fibrosis and assessing their severity, and for predicting treatment response and recurrence of hepatocellular carcinoma, outcomes of liver transplantation recipients, and risk of drug-induced liver injury. Herein, we aim to comprehensively summarize the current evidence regarding diagnostic, prognostic, and/or therapeutic role of AI in these common liver diseases.

Treatment of liver fibrosis: Past, current, and future

Liver fibrosis accompanies the progression of chronic liver diseases independent of etiologies, such as hepatitis viral infection, alcohol consumption, and metabolic-associated fatty liver disease. It is commonly associated with liver injury, inflammation, and cell death. Liver fibrosis is characterized by abnormal accumulation of extracellular matrix components that are expressed by liver myofibroblasts such as collagens and alpha-smooth actin proteins. Activated hepatic stellate cells contribute to the major population of myofibroblasts. Many treatments for liver fibrosis have been investigated in clinical trials, including dietary supplementation (e.g., vitamin C), biological treatment (e.g., simtuzumab), drug (e.g., pegbelfermin and natural herbs), genetic regulation (e.g., non-coding RNAs), and transplantation of stem cells (e.g., hematopoietic stem cells). However, none of these treatments has been approved by Food and Drug Administration. The treatment efficacy can be evaluated by histological staining methods, imaging methods, and serum biomarkers, as well as fibrosis scoring systems, such as fibrosis-4 index, aspartate aminotransferase to platelet ratio, and non-alcoholic fatty liver disease fibrosis score. Furthermore, the reverse of liver fibrosis is slowly and frequently impossible for advanced fibrosis or cirrhosis. To avoid the life-threatening stage of liver fibrosis, anti-fibrotic treatments, especially for combined behavior prevention, biological treatment, drugs or herb medicines, and dietary regulation are needed. This review summarizes the past studies and current and future treatments for liver fibrosis.

Machine learning models are superior to noninvasive tests in identifying clinically significant stages of NAFLD and NAFLD-related cirrhosis

BACKGROUND AND AIMS

We assessed the performance of machine learning (ML) models in identifying clinically significant NAFLD-associated liver fibrosis and cirrhosis.

APPROACH AND RESULTS

We implemented ML models including logistic regression (LR), random forest (RF), and artificial neural network to predict histological stages of fibrosis using 17 demographic/clinical features in 1370 patients with NAFLD who underwent liver biopsy, FibroScan, and labs within a 6-month period at multiple U.S. centers. Histological stages of fibrosis (≥F2, ≥F3, and F4) were predicted using ML, FibroScan liver stiffness measurements, and Fibrosis-4 index (FIB-4). NASH with significant fibrosis (NAS ≥ 4 + ≥F2) was assessed using ML, FibroScan-AST (FAST) score, FIB-4, and NAFLD fibrosis score (NFS). We used 80% of the cohort to train and 20% to test the ML models. For ≥F2, ≥F3, F4, and NASH + NAS ≥ 4 + ≥F2, all ML models, especially RF, had primarily higher accuracy and AUC compared with FibroScan, FIB-4, FAST, and NFS. AUC for RF versus FibroScan and FIB-4 for ≥F2, ≥F3, and F4 were (0.86 vs. 0.81, 0.78), (0.89 vs. 0.83, 0.82), and (0.89 vs. 0.86, 0.85), respectively. AUC for RF versus FAST, FIB-4, and NFS for NASH + NAS ≥ 4 + ≥F2 were (0.80 vs. 0.77, 0.66, 0.63). For NASH + NAS ≥ 4 + ≥F2, all ML models had lower/similar percentages within the indeterminate zone compared with FIB-4 and NFS. Overall, ML models performed better in sensitivity, specificity, positive predictive value, and negative predictive value compared with traditional noninvasive tests.

CONCLUSIONS

ML models performed better overall than FibroScan, FIB-4, FAST, and NFS. ML could be an effective tool for identifying clinically significant liver fibrosis and cirrhosis in patients with NAFLD.

Artificial intelligence models for the diagnosis and management of liver diseases

With the development of more advanced methods for the diagnosis and treatment of diseases, the data required for medical care are becoming complex, and misinterpretation of information due to human error may result in serious consequences. Human error can be avoided with the support of artificial intelligence (AI). AI models trained with various medical data for diagnosis and management of liver diseases have been applied to hepatitis, fatty liver disease, liver cirrhosis, and liver cancer. Some of these models have been reported to outperform human experts in terms of performance, indicating their potential for supporting clinical practice given their high-speed output. This paper summarizes the recent advances in AI for liver disease and introduces the AI-aided diagnosis of liver tumors using B-mode ultrasonography.

Registered Trials of Artificial Intelligence Conducted on Chronic Liver Disease: A Cross-Sectional Study on ClinicalTrials.gov

Background

Artificial intelligence (AI) has been widely applied in the diagnosis and therapy of chronic liver disease (CLD), but there is currently little insight into the trials registered on ClinicalTrials.gov. Thus, this cross-sectional study was focused on analyzing the progress in the use of AI in CLD.

Methods

Registered trials of AI applied in CLD on ClinicalTrials.gov were searched firstly. All available information was downloaded to Excel (Microsoft Excel, Rong, Rong, China), and duplicates were removed. We extracted the data of the included trials, then analyzed the characteristics of them finally.

Results

Up to the 27th of May 2021, 6835 trials were identified following an initial search, and 20 registered trials were included after screening for inclusion and exclusion criteria. Among those trials, hepatocellular carcinoma (HCC, 40.0%) and nonalcoholic fatty liver disease (NAFLD, 20.0%) were the most widely applied CLDs for AI. Trials started in 2013 until 2021, with 17 trials (85%) registered after 2016. There was a large trend in trial enrolment, with 40% of them including samples more than 500. Five trials (25%) have been completed, but only one of these had available results. The most frequent sponsors and collaborators were both hospitals at 55%, followed by universities at 35% and institutes at 11%, respectively. Of the 20 trials included, 35% (7 trials) were interventional trials and 65% (13 trials) were observational trials. Among 7 interventional trials, most trials were for diagnosis purpose (42.86%, 3 trials); 4 trials (57.14%) were randomized; 3 trials (42.86%) applied behavioral intervention, 1 trial (14.29%) was in device intervention, 2 trials (28.57%) were in diagnostic test, and 1 trial intervention was unknown. Among 13 observational trials, 8 (61.54%) were cohort studies; 6 (46.15%) were prospective studies, 4 (30.77%) were retrospective studies, 2 (15.38%) were cross-sectional studies, and 1 (7.69%) did not involve a temporal perspective.

Conclusion

The study is the first to focus on AI registration trials in CLD, which will aid relevant scholars in understanding the current state of the subject. This study demonstrates that additional research on AI used in the diagnosis and treatment of CLD is required, and timely publication of accessible results from registered trials is essential.

Noninvasive staging of liver fibrosis: review of current quantitative CT and MRI-based techniques

Liver fibrosis features excessive protein accumulation in the liver interstitial space resulting from repeated tissue injury due to chronic liver disease. Liver fibrosis eventually proceeds to cirrhosis and associated complications. So, early diagnosis and staging of liver fibrosis are of vital importance for clinical treatment. Liver biopsy remains the gold standard for the diagnosing and staging of fibrosis, but it is suboptimal due to various limitations. Recently, efforts have been made to migrate toward noninvasive techniques for assessing liver fibrosis. CT is relatively easy to perform, relatively standardized for different scanners, and does not require additional hardware in liver fibrosis staging. MRI is frequently performed to characterize indeterminate liver lesions. Because it does not use ionizing radiation and features high image contrast, its role has increased in the staging of liver fibrosis. More recently, several studies on liver fibrosis staging using deep learning algorithms in CT or MRI have been proposed and have shown meaningful results. In this review, we summarize the basic concept, diagnostic performance, and advantages and limitations of each technique to noninvasively stage liver fibrosis.

Simultaneous liver steatosis, fibrosis and iron deposition quantification with mDixon quant based on radiomics analysis in a rabbit model

PURPOSE

To evaluate the feasibility of simultaneous quantification of liver fibrosis, liver steatosis and abnormal iron deposition using mDixon Quant based on radiomics analysis, and to eliminate the interference among different histopathologic features.

METHODS

One hundred and twenty rabbits that were administered CCl4 for 4-16 weeks and a cholesterol rich diet for the initial 4 weeks in the experimental group and 20 rabbits in the control group were examined using mDixon. Radiomics features of the whole liver were extracted from PDFF and R2* and radiomics models for discriminating steatosis: S0-S1 vs. S2-S4, fibrosis: F0-F2 vs. F3-F4 and iron deposition: normal vs. abnormal were constructed respectively and evaluated using receiver operating characteristic (ROC) curves with the histopathological results as reference standard. Combined corrected models merging the radscore and the other two histopathologic features were evaluated using multiple logistic regression analyses and compared with radiomics models.

RESULTS

The area under the ROC curve (AUC) of the radiomics model with PDFF features was 0.886 and 0.843 in the training and the test set, respectively, for the diagnosis of liver steatosis grade S0-1 and S2-S4. The radiomics model based on R2* features were 0.815 and 0.801 for distinguishing F0-F2 and F3-F4 and 0.831 and 0.738 for discriminating abnormal iron deposition in the training and test set, respectively. The corrected model for liver steatosis and fibrosis (0.944 and 0.912 in the test set) outperformed the radiomics models by eliminating the interference of histopathologic features(P < 0.05), but had comparable diagnostic performance for abnormal iron deposition(P > 0.05).

CONCLUSIONS

It is feasible for mDixon to simultaneously quantify whole liver steatosis, fibrosis and iron deposition based on radiomics analysis. It is valuable to minimize the interference of different pathological features for the assessment of liver steatosis and fibrosis.

Artificial intelligence in the diagnosis of cirrhosis and portal hypertension

Clinically significant portal hypertension is associated with an increased risk of developing gastroesophageal varices and hepatic decompensation. Hepatic venous pressure gradient measurement and esophagogastroduodenoscopy are the gold-standard methods for assessing clinically significant portal hypertension and gastroesophageal varices, respectively. However, invasiveness, cost, and feasibility limit their widespread use, especially if repeated and serial evaluations are required to assess the efficacy of pharmacotherapy. Artificial intelligence describes a range of techniques that allow machines to perform tasks typically thought to require human reasoning and problem-solving skills. Artificial intelligence has made great strides in the field of medicine, and is also involved in portal hypertension diagnosis. Artificial intelligence tools will potentially transform our practice by leveraging massive amounts of data to personalize care to the right patient, in the right amount, at the right time. This review focuses on the recent advances in artificial intelligence for the noninvasive diagnosis of portal hypertension and gastroesophageal varices and monitoring of risk assessment of its complications in clinical practice.

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.

Cardiac-induced liver deformation as a measure of liver stiffness using dynamic imaging without magnetization tagging-preclinical proof-of-concept, clinical translation, reproducibility and feasibility in patients with cirrhosis

PURPOSE

MR elastography and magnetization-tagging use liver stiffness (LS) measurements to diagnose fibrosis but require physical drivers, specialist sequences and post-processing. Here we evaluate non-rigid registration of dynamic two-dimensional cine MRI images to measure cardiac-induced liver deformation (LD) as a measure of LS by (i) assessing preclinical proof-of-concept, (ii) clinical reproducibility and inter-reader variability, (iii) the effects of hepatic hemodynamic changes and (iv) feasibility in patients with cirrhosis.

METHODS

Sprague-Dawley rats (n = 21 bile duct ligated (BDL), n = 17 sham-operated controls) and fasted patients with liver cirrhosis (n = 11) and healthy volunteers (HVs, n = 10) underwent spoiled gradient-echo short-axis cardiac cine MRI studies at 9.4 T (rodents) and 3.0 T (humans). LD measurements were obtained from intrahepatic sub-cardiac regions-of-interest close to the diaphragmatic margin. One-week reproducibility and prandial stress induced hemodynamic changes were assessed in healthy volunteers.

RESULTS

Normalized LD was higher in BDL (1.304 ± 0.062) compared with sham-operated rats (1.058 ± 0.045, P = 0.0031). HV seven-day reproducibility Bland-Altman (BA) limits-of-agreement (LoAs) were ± 0.028 a.u. and inter-reader variability BA LoAs were ± 0.030 a.u. Post-prandial LD increases were non-significant (+ 0.0083 ± 0.0076 a.u., P = 0.3028) and uncorrelated with PV flow changes (r = 0.42, p = 0.2219). LD measurements successfully obtained from all patients were not significantly higher in cirrhotics (0.102 ± 0.0099 a.u.) compared with HVs (0.080 ± 0.0063 a.u., P = 0.0847).

CONCLUSION

Cardiac-induced LD is a conceptually reasonable approach from preclinical studies, measurements demonstrate good reproducibility and inter-reader variability, are less likely to be affected by hepatic hemodynamic changes and are feasible in patients with cirrhosis.

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 in gastroenterology and hepatology: Status and challenges

Originally proposed by John McCarthy in 1955, artificial intelligence (AI) has achieved a breakthrough and revolutionized the processing methods of clinical medicine with the increasing workloads of medical records and digital images. Doctors are paying attention to AI technologies for various diseases in the fields of gastroenterology and hepatology. This review will illustrate AI technology procedures for medical image analysis, including data processing, model establishment, and model validation. Furthermore, we will summarize AI applications in endoscopy, radiology, and pathology, such as detecting and evaluating lesions, facilitating treatment, and predicting treatment response and prognosis with excellent model performance. The current challenges for AI in clinical application include potential inherent bias in retrospective studies that requires larger samples for validation, ethics and legal concerns, and the incomprehensibility of the output results. Therefore, doctors and researchers should cooperate to address the current challenges and carry out further investigations to develop more accurate AI tools for improved clinical applications.

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

Background

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

Methods

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

Results

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

Conclusions

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

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