A Systematic Machine Learning Based Approach for the Diagnosis of Non-Alcoholic Fatty Liver Disease Risk and Progression

Sajida Perveen, Muhammad Shahbaz, Karim Keshavjee, Aziz Guergachi
2018 Scientific Reports  
Prevention and diagnosis of NAFLD is an ongoing area of interest in the healthcare community. Screening is complicated by the fact that the accuracy of noninvasive testing lacks specificity and sensitivity to make and stage the diagnosis. Currently no non-invasive ATP III criteria based prediction method is available to diagnose NAFLD risk. Firstly, the objective of this research is to develop machine learning based method in order to identify individuals at an increased risk of developing
more » ... using risk factors of ATP III clinical criteria updated in 2005 for Metabolic Syndrome (MetS). Secondly, to validate the relative ability of quantitative score defined by Italian Association for the Study of the Liver (IASF) and guideline explicitly defined for the Canadian population based on triglyceride thresholds to predict NAFLD risk. We proposed a Decision Tree based method to evaluate the risk of developing NAFLD and its progression in the Canadian population, using Electronic Medical Records (EMRs) by exploring novel risk factors for NAFLD. Our results show proposed method could potentially help physicians make more informed choices about their management of patients with NAFLD. Employing the proposed application in ordinary medical checkup is expected to lessen healthcare expenditures compared with administering additional complicated test. NAFLD is a common clinico-pathologic entity that includes a wide spectrum of liver disorders. This ranges from simple steatosis (excessive fat accumulation in liver) to steatohepatitis (liver cell injury and inflammation), advanced fibrosis and rarely, progression to cirrhosis and hepatocellular carcinoma. It is marked by hepatic triglyceride (TRG) accumulation in liver parenchyma that adds to liver weight by at least 5%, however, it is not caused by consumption of alcohol 1,2 . NAFLD prevalence is increasing rapidly. This increase is quite noteworthy in western countries. According to Souza et al. 3 the prevalence of NAFLD is estimated at 45% in Hispanic-Americans, 33% in European-Americans and 24% in African-American. Other studies show that it can affect up to 30% of the general population 4 . Its relative prevalence is estimated to be 69% among individuals with type 2 diabetes mellitus/glucose intolerance 5 when diagnosed by ultrasonography, 87% when diagnosed using biopsy or magnetic resonance imaging 6,7 . The literal pervasiveness of NAFLD still remains unidentified due to heterogeneity in diagnosis, the population under consideration and the degree of diversity across various factors associated therewith 8 . Although the pathogenic mechanism of NAFLD is incompletely understood, the majority of NAFLD patients remains oblivious of their diagnosis until some major complications are encountered or it is diagnosed during tests carried out for some other reasons 9 . NAFLD bears bidirectional association with Metabolic Syndrome (MetS) 10 . MetS is a cluster of risk factors that significantly exposes an individual to coronary heart disease, diabetes mellitus, endocrine-metabolic diseases and chronic renal failure 7,11,12 . Hence a space is available to make use of these factors for diagnosis of NAFLD risks. In 2005, the clinical criteria Adult Treatment Panel III (ATP III) were
doi:10.1038/s41598-018-20166-x pmid:29391513 pmcid:PMC5794753 fatcat:m4c27mk5p5balds6bw2ltgwwfq