Laboratory parameter‐based machine learning model for excluding non‐alcoholic fatty liver disease (<scp>NAFLD</scp>) in the general population

Yip, Terry Cheuk‐Fung; J., A.; Wong, Vincent Wai‐Sun; Tse, Yee Kit; Chan, Henry Lik‐Yuen; Yuen, Pong C.; Wong, Grace Lai–Hung

doi:10.1111/apt.14172

Cited by 142 publications

(148 citation statements)

References 28 publications

Supporting

Mentioning

146

Contrasting

Unclassified

Order By: Relevance

“…On multivariable analysis, adjusted hazard ratios (aHR) with 95% CIs were estimated with Cox proportional hazards regression, and Fine‐Gray proportional sub‐distribution hazards regression with adjustment of competing risk . Since the elastic net regularisation technique constrained sparsity on the 29 parameters for prediction, we employed this technique to select the related small subset of parameters automatically. Compared with multivariate analysis followed by stepwise parameter selection, elastic net regularisation selects variables and optimizes model performance simultaneously under a continuous pathway.…”

Section: Methodsmentioning

confidence: 99%

“…Machine learning is a comprehensive tool arisen in recent years for model development, allows direct selection of predicting parameters among all available parameters without subjective preselection, and maximises data use while minimises bias. Machine learning is used to improve the diagnosis of non‐alcoholic fatty liver disease from electronic medical record data . In this study, we aimed to develop a novel clinical and laboratory parameter‐based prediction model using machine learning algorithm to predict recurrent ulcer bleeding in patients with a history of H. pylori –negative idiopathic bleeding ulcers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning model to predict recurrent ulcer bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding

Wong

Andy

Deng

et al. 2019

Aliment Pharmacol Ther

Self Cite

View full text Add to dashboard Cite

Background: Patients with a history of Helicobacter pylori-negative idiopathic bleeding ulcers have an increased risk of recurring ulcer complications. Aim: To build a machine learning model to identify patients at high risk for recurrent ulcer bleeding. Methods: Data from a retrospective cohort of 22 854 patients (training cohort) diagnosed with peptic ulcer disease in 2007-2016 were analysed to build a model (IPU-ML) to predict recurrent ulcer bleeding. We tested the IPU-ML in all patients with a diagnosis of gastrointestinal bleeding (n = 1265) in 2008-2015 from a different catchment population (independent validation cohort). Any co-morbid conditions which had occurred in >1% of study population were eligible as predictors. Results: Recurrent ulcer bleeding developed in 4772 patients (19.5%) in the training cohort, during a median follow-up period of 2.7 years. IPU-ML model built on six parameters (age, baseline haemoglobin, and presence of gastric ulcer, gastrointestinal diseases, malignancies, and infections) identified patients with bleeding recurrence within 1 year with an area under the receiver operating characteristic curve (AUROC) of 0.648. When we set the IPU-ML cutoff value at 0.20, 27.5% of patients were classified as high risk for rebleeding with a sensitivity of 41.4%, specificity of 74.6%, and a negative predictive value of 91.1%. In the validation cohort, the IPU-ML identified patients with a recurrence ulcer bleeding within 1 year with an AUROC of 0.775, and 84.3% of overall accuracy. Conclusion:We developed a machine-learning model to identify those patients with a history of idiopathic gastroduodenal ulcer bleeding who are not at high risk for recurrent ulcer bleeding.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Machine learning model to predict recurrent ulcer bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding

Wong

Andy

Deng

et al. 2019

Aliment Pharmacol Ther

Self Cite

View full text Add to dashboard Cite

show abstract

“…ML models have been used to detect fibrosis or cirrhosis secondary to different liver disease etiologies based on blood tests. (34)(35)(36)(37)(38)(39)(40)(41) These studies have predicted the risk of cirrhosis in order to reduce the overall number of liver biopsies done on patients with known risk factors. Diagnostic performance of these models has varied depending on the etiology of liver disease and study size.…”

Section: In Liver Diseasesmentioning

confidence: 99%

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

et al. 2020

View full text Add to dashboard Cite

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.

show abstract

“…AP&T recently published an interesting study describing the development of a model to predict steatosis using routine clinical and laboratory parameters . The study included 922 individuals, selected randomly from the general population of Hong Kong, who had a comprehensive assessment for NAFLD, including MR spectroscopy.…”

mentioning

confidence: 99%

Editorial: progress towards a simple tool for screening for hepatic steatosis in the general population

Gallacher

McPherson

2017

Aliment Pharmacol Ther

View full text Add to dashboard Cite

Non-alcoholic fatty liver disease (NAFLD) is a significant cause of endstage liver disease and hepatocellular carcinoma. Despite, studies showing that 20%-40% of the general population have hepatic steatosis, it often goes unrecognised in primary care.1 Current guidelines do not advocate population screening for NAFLD, 2,3 largely due to the lack of a simple, cost-effective test for steatosis, and the fact there are relatively few treatment options available. However, several drugs for NAFLD are in advanced clinical trials and it is likely that some will be licensed soon. 4 Therefore, development of a simple test to identify subjects with steatosis in the population would be advantageous.AP&T recently published an interesting study describing the development of a model to predict steatosis using routine clinical and laboratory parameters. 5 The study included 922 individuals, selected randomly from the general population of Hong Kong, who had a comprehensive assessment for NAFLD, including MR spectroscopy. Using advanced statistical modelling techniques four models were developed to predict the presence of steatosis (>5%), which all had good diagnostic precision (area under receiver operator curve[AUROC] 0.87-0.9). The authors concluded that the "NAFLD ridge score" offered the best balance between efficacy and simplicity. This model included 6 parameters (serum triglycerides, alanine aminotransferase, high density lipoprotein cholesterol, haemoglobin A1c, white cell count and the presence of hypertension) that are routinely available for individuals undergoing medical checkups. The AUROCs to predict steatosis for this score were 0.87 and 0.88 in the training and validation cohorts respectively. Using the lower cut-off (0.24) the negative predictive value for steatosis was 96%, suggesting the score is reliable in excluding steatosis. Using the upper cut-off (0.44) the positive predictive value for steatosis was 70%. Therefore, the main value of this score would be to exclude steatosis, similar to the use of available simple non-invasive fibrosis scores. 6,7The NAFLD ridge score has advantages over current scores to predict steatosis (e.g. fatty liver index) as it does not require input of anthropometric measures, which are not always available. 8 Given its simplicity the NAFLD ridge score could be used to screen individuals periodically to identify/exclude NAFLD.Although the study was well conducted, it has few limitations that are worthy of discussion. Firstly, this study was conducted in aChinese population so the findings need to be confirmed in other ethnicities. Secondly, the impact of alcohol consumption on the performance of the score requires further study. In addition, the effect of treatment of diabetes and dyslipidaemia on the performance of the score was not addressed.In our view, the NAFLD ridge score may also benefit from a single cut-off with optimum sensitivity and specificity to predict steatosis. While this would reduce its overall accuracy, practically it would simplify use of the tool and...

show abstract

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

Abstract: NAFLD ridge score is a simple and robust reference comparable to existing NAFLD scores to exclude NAFLD patients in epidemiological studies.

Cited by 142 publications

References 28 publications

Machine learning model to predict recurrent ulcer bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding

Machine learning model to predict recurrent ulcer bleeding in patients with history of idiopathic gastroduodenal ulcer bleeding

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

Editorial: progress towards a simple tool for screening for hepatic steatosis in the general population

Contact Info

Product

Resources

About