Relevant Features in Nonalcoholic Steatohepatitis Determined Using Machine Learning for Feature Selection.
Journal:
Metabolic syndrome and related disorders
Published Date:
Nov 1, 2019
Abstract
We investigated the prevalence and the most relevant features of nonalcoholic steatohepatitis (NASH), a stage of nonalcoholic fatty liver disease, (NAFLD) in which the inflammation of hepatocytes can lead to increased cardiovascular risk, liver fibrosis, cirrhosis, and the need for liver transplant. We analyzed data from 2239 hypertensive patients using descriptive statistics and supervised machine learning algorithms, including the least absolute shrinkage and selection operator and random forest classifier, to select the most relevant features of NASH. The prevalence of NASH among our hypertensive patients was 11.3%. In univariate analyses, it was associated with metabolic syndrome, type 2 diabetes, insulin resistance, and dyslipidemia. Ferritin and serum insulin were the most relevant features in the final model, with a sensitivity of 70%, specificity of 79%, and area under the curve of 0.79. Ferritin and insulin are significant predictors of NASH. Clinicians may use these to better assess cardiovascular risk and provide better management to hypertensive patients with NASH. Machine-learning algorithms may help health care providers make decisions.
Authors
Keywords
Adult
Aged
Algorithms
Cross-Sectional Studies
Decision Making
Diabetes Mellitus, Type 2
Disease Progression
Female
Humans
Hypertension
Liver Cirrhosis
Machine Learning
Male
Metabolic Syndrome
Middle Aged
Non-alcoholic Fatty Liver Disease
Prevalence
Prognosis
Retrospective Studies
Sensitivity and Specificity
Severity of Illness Index