Predicting Liver Patients using Artificial Neural Network

Download Edit this record How to cite View on PhilPapers
Abstract
Liver diagnosis at an early stage is essential for enhanced handling. Precise classification is required for automatic recognition of disease from data samples (utilizing data mining for classification of liver patients from healthy ones). In this study, an artificial neural network model was designed and developed using JustNN Tool for predicting weather a person is a liver patient or not based on a dataset for liver patients. The main factors for input variables are: Age, Gender, Total Bilirubin, Direct Bilirubin, Alkphos Alkaline Phosphotase. Sgpt Alamine Aminotransferase, Sgot Aspartate Aminotransferase, Total Protiens, Albumin, Albumin and Globulin Ratio, and the output variable: Status. The dataset used for training are the data published in the literature for various 583 liver patients. The model was trained and validated, most important factors affecting Status of liver patient identified, and the accuracy for the validation was 99.00%.
Keywords
Categories
(categorize this paper)
PhilPapers/Archive ID
BARPLP
Upload history
Archival date: 2019-11-18
View other versions
Added to PP index
2019-11-18

Total views
423 ( #14,204 of 2,440,213 )

Recent downloads (6 months)
153 ( #3,658 of 2,440,213 )

How can I increase my downloads?

Downloads since first upload
This graph includes both downloads from PhilArchive and clicks on external links on PhilPapers.