Gradient Boosting Model for Non-Invasive Liver Fibrosis Staging using Plasma Metabolomics and Clinical Biochemistry

• Israa M Shamkh

  Chief Computational Chemistry Department EBO Bio Solution Company, London, EC1V2NX, United Kingdom

  Author

• Ahmed Hassen Shntaif

  Assist Professor of Chemistry Department of Chemistry, University of Babylon, Alhilla, Iraq

  Author

• Ahmed Helmy

  Student At Department Of Biotechnology Faculty Of Agriculture Ain Shams University Cairo Egypt

  Author

• Ihosvany Camps

  Full Professor. Department of Physics, Federal University of Alfenas, Alfenas, MG, Brazil

  Author

A serious worldwide health issue linked to the development of chronic liver disorders of different etiologies is liver fibrosis. Making informed judgements about therapy and prognosis requires accurate non-invasive evaluation of the stage of liver fibrosis. Although liver biopsy is still the gold standard, it has drawbacks such invasiveness, unpredictability in sample, and potential for consequences. We selected genomic datasets from genomic library which the data of 30 individuals with chronic liver disease. The mean age of the 18 men and 12 females was 48.9 ± 12.7 years. Alcohol (n = 2), hepatitis B (n = 12), hepatitis C (n = 11), non-alcoholic fatty liver disease (n = 5), and other causes (n = 5) were among the etiologies of liver disease.250 individuals had their plasma samples taken while fasting. Ten percent of each sample's plasma was pooled to create quality control samples. An AUROC of 0.93 (95% CI=0.90-0.96, p<0.0001) for F≥2 discrimination was attained by the final optimized GBM model in the training set (n= 250). The model showed good discriminating power with an AUROC of 0.92 (95% CI=0.86-0.98, p<0.0001) when applied to the validation set (n = 50).

• PDF

• Biotechnology Section
• Computational Science Section

This work is licensed under a Creative Commons Attribution 4.0 International License.