Digital image analysis of ultrasound images using machine learning to diagnose pediatric nonalcoholic fatty liver disease

Neural Network Classification Algorithm Based on Self-attention Mechanism and Ensemble Learning for MASLD Ultrasound Images

BACKGROUND

Ultrasound image examination has become the preferred choice for diagnosing metabolic dysfunction-associated steatotic liver disease (MASLD) due to its non-invasive nature. Computer-aided diagnosis (CAD) technology can assist doctors in avoiding deviations in the detection and classification of MASLD.

METHOD

We propose a hybrid model that integrates the pre-trained VGG16 network with an attention mechanism and a stacking ensemble learning model, which is capable of multi-scale feature aggregation based on the self-attention mechanism and multi-classification model fusion (Logistic regression, random forest, support vector machine) based on stacking ensemble learning. The proposed hybrid method achieves four classifications of normal, mild, moderate, and severe fatty liver based on ultrasound images.

RESULT AND CONCLUSION

Our proposed hybrid model reaches an accuracy of 91.34% and exhibits superior robustness against interference, which is better than traditional neural network algorithms. Experimental results show that, compared with the pre-trained VGG16 model, adding the self-attention mechanism improves the accuracy by 3.02%. Using the stacking ensemble learning model as a classifier further increases the accuracy to 91.34%, exceeding any single classifier such as LR (89.86%) and SVM (90.34%) and RF (90.73%). The proposed hybrid method can effectively improve the efficiency and accuracy of MASLD ultrasound image detection.

Efficacy of Ultrasound for the Detection of Possible Fatty Liver Disease in Children

Pediatric MASLD (previously referred to as NAFLD) incidence has continued to rise along with the obesity pandemic. Pediatric MASLD increases the risk of liver fibrosis and cirrhosis in adulthood. Early detection and intervention can prevent and reduce complications. Liver biopsy remains the gold standard for diagnosis, although imaging modalities are increasingly being used. We performed a retrospective study of 202 children seen in a pediatric gastroenterology clinic with a complaint of abdominal pain, elevated liver enzymes or MASLD, or a combination of the three to evaluate screening methods for MASLD. A total of 134 of the 202 patients included in the study underwent laboratory testing and abdominal ultrasound. Ultrasound images were reviewed with attention to liver size and echotexture by a fellowship-trained pediatric radiologist for liver size and echotexture. Overall, 76.2% of the initial radiology reports correctly identified hepatomegaly based on age and 75.4% of the initial radiology reports correctly described hepatic echogenicity that was consistent with increased hepatic fat deposition. Use of screening ultrasound in concert with other clinical evaluations can be helpful to identify children at risk of MASLD. Utilizing ranges for liver span according to age can help to diagnose hepatomegaly, and understanding how to identify hepatic echogenicity is important for identifying possible hepatic steatosis.

Prediction of Fatty Liver Disease in a Chinese Population Using Machine-Learning Algorithms

BACKGROUND

Fatty liver disease (FLD) is an important risk factor for liver cancer and cardiovascular disease and can lead to significant social and economic burden. However, there is currently no nationwide epidemiological survey for FLD in China, making early FLD screening crucial for the Chinese population. Unfortunately, liver biopsy and abdominal ultrasound, the preferred methods for FLD diagnosis, are not practical for primary medical institutions. Therefore, the aim of this study was to develop machine learning (ML) models for screening individuals at high risk of FLD, and to provide a new perspective on early FLD diagnosis.

METHODS

This study included a total of 30,574 individuals between the ages of 18 and 70 who completed abdominal ultrasound and the related clinical examinations. Among them, 3474 individuals were diagnosed with FLD by abdominal ultrasound. We used 11 indicators to build eight classification models to predict FLD. The model prediction ability was evaluated by the area under the curve, sensitivity, specificity, positive predictive value, negative predictive value, and kappa value. Feature importance analysis was assessed by Shapley value or root mean square error loss after permutations.

RESULTS

Among the eight ML models, the prediction accuracy of the extreme gradient boosting (XGBoost) model was highest at 89.77%. By feature importance analysis, we found that the body mass index, triglyceride, and alanine aminotransferase play important roles in FLD prediction.

CONCLUSION

XGBoost improves the efficiency and cost of large-scale FLD screening.

Artificial Intelligence for Detecting and Quantifying Fatty Liver in Ultrasound Images: A Systematic Review

BACKGROUND

Non-alcoholic Fatty Liver Disease (NAFLD) is growing more prevalent worldwide. Although non-invasive diagnostic approaches such as conventional ultrasonography and clinical scoring systems have been proposed as alternatives to liver biopsy, their efficacy has been called into doubt. Artificial Intelligence (AI) is now combined with traditional diagnostic processes to improve the performance of non-invasive approaches.

OBJECTIVE

This study explores how well various AI methods function and perform on ultrasound (US) images to diagnose and quantify non-alcoholic fatty liver disease.

METHODOLOGY

A systematic review was conducted to achieve this objective. Five science bibliographic databases were searched, including PubMed, Association for Computing Machinery ACM Digital Library, Institute of Electrical and Electronics Engineers IEEE Xplore, Scopus, and Google Scholar. Only peer-reviewed English articles, conferences, theses, and book chapters were included. Data from studies were synthesized using narrative methodologies per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria.

RESULTS

Forty-nine studies were included in the systematic review. According to the qualitative analysis, AI significantly enhanced the diagnosis of NAFLD, Non-Alcoholic Steatohepatitis (NASH), and liver fibrosis. In addition, modalities, image acquisition, feature extraction and selection, data management, and classifiers were assessed and compared in terms of performance measures (i.e., accuracy, sensitivity, and specificity).

CONCLUSION

AI-supported systems show potential performance increases in detecting and quantifying steatosis, NASH, and liver fibrosis in NAFLD patients. Before real-world implementation, prospective studies with direct comparisons of AI-assisted modalities and conventional techniques are necessary.

A systematic review of predictive models for hospital-acquired pressure injury using machine learning

AIMS AND OBJECTIVES

To summarize the use of machine learning (ML) for hospital-acquired pressure injury (HAPI) prediction and to systematically assess the performance and construction process of ML models to provide references for establishing high-quality ML predictive models.

BACKGROUND

As an adverse event, HAPI seriously affects patient prognosis and quality of life, and causes unnecessary medical investment. At present, the performance of various scales used to predict HAPIs is still unsatisfactory. As a new statistical tool, ML has been applied to predict HAPIs. However, its performance has varied in different studies; moreover, some deficiencies in the model construction process were observed in each study.

DESIGN

Systematic review.

METHODS

Relevant articles published between 2010-2021 were identified in the PubMed, Web of Science, Scopus, Embase and CINHAL databases. Study selection was performed in accordance with the preferred reporting items for systematic reviews and meta-analysis guidelines. The quality of the included articles was assessed using the prediction model risk of bias assessment tool.

RESULTS

Twenty-three studies out of 1793 articles were considered in this systematic review. The sample size of each study ranged from 149-75353; the prevalence of pressure injuries ranged from 0.5%-49.8%. ML showed good performance for HAPI prediction. However, some deficiencies were observed in terms of data management, data pre-processing and model validation.

CONCLUSIONS

ML, as a powerful decision-making assistance tool, is helpful for the prediction of HAPIs. However, existing studies have been insufficient in terms of data management, data pre-processing and model validation. Future studies should address these issues to establish ML models for HAPI prediction that can be widely used in clinical practice.

RELEVANCE TO CLINICAL PRACTICE

This review highlights that ML is helpful in predicting HAPI; however, in the process of data management, data pre-processing and model validation, some deficiencies still need to be addressed. The ultimate goal of integrating ML into HAPI prediction is to develop a practical clinical decision-making tool. A complete and rigorous model construction process should be followed in future studies to develop high-quality ML models that can be applied in clinical practice.

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

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

Clinical spectrum transition and prediction model of nonalcoholic fatty liver disease in children with obesity

OBJECTIVE

This study aims to outline the clinical characteristics of pediatric NAFLD, as well as establish and validate a prediction model for the disease.

MATERIALS AND METHODS

The retrospective study enrolled 3216 children with obesity from January 2003 to May 2021. They were divided into obese without NAFLD, nonalcoholic fatty liver (NAFL), and nonalcoholic steatohepatitis (NASH) groups. Clinical data were retrieved, and gender and chronologic characteristics were compared between groups. Data from the training set (3036) were assessed using univariate analyses and stepwise multivariate logistic regression, by which a nomogram was developed to estimate the probability of NAFLD. Another 180 cases received additional liver hydrogen proton magnetic resonance spectroscopy (1H-MRS) as a validation set.

RESULTS

The prevalence of NAFLD was higher in males than in females and has increased over the last 19 years. In total, 1915 cases were NAFLD, and the peak onset age was 10-12 years old. Hyperuricemia ranked first in childhood NAFLD comorbidities, followed by dyslipidemia, hypertension, metabolic syndrome (MetS), and dysglycemia. The AUROC of the eight-parameter nomogram, including waist-to-height ratio (WHtR), hip circumference (HC), triglyceride glucose-waist circumference (TyG-WC), alanine aminotransferase (ALT), high-density lipoprotein cholesterol (HDL-C), apolipoprotein A1(ApoA1), insulin sensitivity index [ISI (composite)], and gender, for predicting NAFLD was 0.913 (sensitivity 80.70%, specificity 90.10%). Calibration curves demonstrated a great calibration ability of the model.

CONCLUSION AND RELEVANCE

NAFLD is the most common complication in children with obesity. The nomogram based on anthropometric and laboratory indicators performed well in predicting NAFLD. This can be used as a quick screening tool to assess pediatric NAFLD in children with obesity.

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.