الفهرس | Only 14 pages are availabe for public view |
Abstract Knowing the degree of a disease allows a physician to provide specific treatment to the patient leading to better care. The use of serum markers becomes important to predict liver fibrosis. The aim of this master’s thesis is to predict liver fibrosis stage by constructing decision trees in patients with chronic hepatitis C genotype 4 in Egypt. Also, we examined whether the combination of certain biomarkers could increase the diagnostic accuracy of liver fibrosis assessment. In addition, this study aims to evaluate whether laboratory examinations can be used to determine the rate of liver fibrosis progression in patients chronically infected with hepatitis C virus genotype 4. This paper focuses on the decision tree classification technique which used to build classification models in the form of a tree structure. In this study, we included most previously known indirect markers of liver fibrosis such as FibroTest which used to stage liver fibrosis to compare the results of the non-invasive markers using histology as reference method. We tried to develop a simple surrogate index comprised of routinely available laboratory tests to reflect the histological fibrosis stage. For the analysis, we used all the data for training as well as for testing as follows: we used 88% as train data to develop the decision tree model and the remaining designated 12% were used as test-data to predict Fibrosis Stage. The results indicated that decision trees model was useful and effective to predict liver fibrosis stage at least similar to biopsy and provide a qualitative and quantitative overview for the physician to find the relations between biomarkers panel and Liver Biopsy. Decision trees produced the correct classification in approximately 94% ofthe cases. Keywords: Biochemical Markers. Liver fibrosis. Hepatitis C virus. Data |