Gut Microbiome profile prediction for Nonalcoholic Fatty Liver disease patients based on Artificial Intelligence Techniques. | ||
| Benha Journal of Applied Sciences | ||
| Volume 10, Issue 7, July 2025, Pages 87-95 PDF (720.49 K) | ||
| Document Type: Original Research Papers | ||
| DOI: 10.21608/bjas.2025.371316.1677 | ||
| Authors | ||
| Bahaa Saifalnasr Rabi* 1; Hanim Khtar2; Ali Hafez El-Far3; Reda Albadawy4 | ||
| 1electrical department ,faculty benha of engineering ,benha university ,benha,Egypt.. | ||
| 2Dept. of Pathology, Faculty of Veterinary Medicine, Benha University, Benha, Egypt | ||
| 3Dept. of Biochemistry, Faculty of Veterinary Medicine, Damanhour University, Damanhour, Egypt | ||
| 4Dept. of Gastrointestinal Diseases, Faculty of Medicine, Benha University, Benha, Egypt | ||
| Abstract | ||
| The current conclusions and hypotheses on the roles of the gut microbiome in the pathophysiological understanding of different phenotypes of nonalcoholic Fatty Liver Disease (NAFLD) patients have greatly attracted research in multiple academic and research institutions. In this study, the gut microbiome profiles of different families and genetic categories that live in NAFLD patients were combined and saved in one dataset as a result of fecal and blood sample analysis. There were distinct gut microbiome patterns in the fecal and blood samples of obese individuals with NAFLD compared to lean individuals. Another important feature used in the dataset is body Mass Index (BMI), which is classified into lean (BMI <25 kg/m^2) and overweight (30>BMI >25 kg/m^2). One of the important branches of Artificial Intelligence is Machine Learning (ML).ML techniques have a great contribution to real-world applications. This approach implements Rules between Gut-Micro profiles for NAFLD (Lean obese). Besides the ML prediction model, Random Forest was deployed to predict the Gut-Micro profile for both lean and obese individuals in the case of NAFLD patients. The testing accuracy in the proposed model is more than 99%, which is considered an excellent performance parameter in gut microbiome investigation. | ||
| Keywords | ||
| Bioinformatics; Body Mass Index; Machine Learning; and Random Forest | ||
|
Statistics Article View: 9 PDF Download: 4 |
||
