Predict of Non-alcoholic Fatty Liver Disease Using Anthropometric Analyzing and Obesity Degree by Machine Learning
محل انتشار: اولین کنگره بین المللی هوش مصنوعی در علوم پزشکی
سال انتشار: 1402
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 113
نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
این مقاله در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
AIMS01_022
تاریخ نمایه سازی: 1 مرداد 1402
چکیده مقاله:
Background and aims: Non-alcoholic fatty liver disease (NAFLD) is one of the most importantcomplications of obesity, especially abdominal obesity. Increased visceral fat is the most importantrisk factors for this. Diseases that are caused by the pathogenesis of insulin resistance areclosely related to this disease. The aim of this study is to investigate the effect of obesity degreeand anthropometric changes in predicting liver steatosis and fibrosis based on artificial intelligence.Methods: A cross-sectional study was conducted among ۶۵۰ individuals over the age of ۱۸ withouta history of continuous alcohol consumption and underlying liver disease in two southern andeastern provinces of Iran. Anthropometric and body composition measurements were performedmanually and body composition analyzer In Body ۲۷۰. Hepatic steatosis and fibrosis were determinedusing a Fibro scan. ML methods including k-Nearest Neighbor (kNN), Support Vector Machine(SVM), Radial Basis Function (RBF) SVM, Gaussian Process (GP), Random Forest (RF),Neural Network (NN), Daboost and Naïve Bayes were examined for model performance and toidentify anthropometric and body composition predictors of fatty liver disease.Results: RF generated the most accurate model for fatty liver (presence of any stage), steatosisstages and fibrosis stages with ۸۲%, ۵۲% and ۵۷% accuracy, respectively. Abdomen circumference,obesity degree, Waist to hip ratio, Waist circumference, trunk fat and body mass index wereamong the most important variables contributing to fatty liver disease.Conclusion: ML-based prediction of NAFLD using obesity degree and visceral obesity rate andeffective anthropometric data. It can help physicians at any level of health in early diagnosis of thedisease and prevention of the progression of the disease towards the occurrence of fibrotic liverand cirrhosis complications.
کلیدواژه ها:
نویسندگان
Farkhoudeh Razmpour
Hormozgan Medical University, Hormozgan, Iran
Seyedeh Aynaz Mousavi Sani
Hormozgan Medical University, Hormozgan, Iran
Ghasem Sadeghi Bajestani
Hormozgan Medical University, Hormozgan, Iran.
Mahdiyeh Razm Pour
Hormozgan Medical University, Hormozgan, Iran