An Empirical Comparative Analysis Using Machine Learning Techniques for Liver Disease Prediction

An Empirical Comparative Analysis Using Machine Learning Techniques for Liver Disease Prediction

Mohammed Alghobiri, Hikmat Ullah Khan, Ahsan Mahmood
DOI: 10.4018/IJHISI.20211001.oa10
Article PDF Download
Open access articles are freely available for download

Abstract

The human liver is one of the major organs in the body and liver disease can cause many problems in human live. Due to the increase in liver disease, various data mining techniques are proposed by the researchers to predict the liver disease. These techniques are improving day by day in order to predict and diagnose the liver disease in human. In this paper, real-world liver disease dataset is incorporated for diagnosing liver disease in human body. For this purpose, feature selection models are used to select a number of features that best are the most important feature to diagnose the liver disease. After selecting features and splitting data for training and testing, different classification algorithms in terms of naïve Bayes, supervised vector machine, decision tree, k near neighbor and logistic regression models to diagnose the liver disease in human body. The results are cross-validated by tenfold cross validation methods and achieve an accuracy as good as 93%.
Article Preview
Top

1. Introduction

Liver is the largest organ of the body. Liver helps the body digesting the food, it store energy and removes poisons (Imenda, 2015). The failure of the liver occurs when the major parts of the liver are damaged and liver stop functioning. This condition is life threatening and it requires urgent medical care (Cressman et al., 1996). Although the condition of the liver occurs gradually due to different reasons, a rarer liver condition is called acute liver failure that is very hard to detect on the initial stage (Alonso & Squires, 2017). There are many symptoms of liver failure and liver diseases which can vary although the most common symptoms are legs swelling, easy bruising, change in stool color, change in urine color, and yellowing eyes (Mochida et al., 2018). Although these are the most common symptoms of the liver diseases, there are some cases when there are no symptoms and it is not possible to diagnose the liver disease. Among the other liver diseases, Liver cancer is the second most common cause of cancer death worldwide, preceded only by lung cancer. There were seven hundred thousand deaths due to liver cancer in 2012 (Torre et al., 2015).

The major reason of with widespread of the liver infection is because of the deskbound lifestyle, alcoholic consumption and smoking. Due to these things, the liver infections is continuously increasing at a very rapid rate. The types of liver diseases including diseases caused by viruses, such as hepatitis A, hepatitis B, and hepatitis C. Diseases caused by drugs, poisons, or too much alcohol(Cheng et al., 2016). Examples include fatty liver disease and cirrhosis. Inherited diseases, such as hemochromatosis and Wilson disease. Liver function test of LFT are used to test the proper function of the liver. These tests are based on a number of steps that normally identify the reason of liver disease and try to treat the liver infection (Yap & Aw, 2010).

In the recent years, due to the availability of the mobile and other devices, the liver infection is easily detected. These smart devices are able to diagnose the liver diseases very easily through the help of different intelligent models and are able to perform much better. The performance of these devices are accelerated by using different classification systems that distinguish between one type and another. These classifier systems are trained by using machine learning models and are able to perform much better if proper machine learning models are incorporated. As the current age of artificial intelligence make it easy to explore the large amount of data, the integration of the previous data into such devices helps in identifying the liver diseases very easily.

Features selection techniques are usually divided into three categories including filter based techniques, wrapper based techniques and embedded methods of feature selection. While the classification algorithms are supervised, semi supervised or unsupervised. These algorithms are based upon different learning mechanism. In supervised machine learning methods, the labels are defined as the results are known. These methods help in better learning as in this method, the dataset is divided into two parts, for training and testing. In the unsupervised learning models, the label patterns are hidden and there are no available labels. These are more complex data learning models with much complex cases. Many researchers use machine learning techniques to solve other problems (Ayesha, Noor, Ramzan, Khan, & Shoaib, 2017; Khan & Daud, 2017; “Using machine learning techniques for subjectivity analysis based on lexical and non-lexical features,” n.d.) by using feature selection and other methods (Ishfaq, Khan, & Iqbal, 2016) (Khan & Daud, 2017). These tools are also helpful in classification (Khan, 2017) and information retrieval (A. Mahmood, Khan, Zahoor-ur-Rehman, & Khan, 2017; Ahsan Mahmood, Khan, Alarfaj, Ramzan, & Ilyas, 2018).

Complete Article List

Search this Journal:
Reset
Volume 19: 1 Issue (2024)
Volume 18: 1 Issue (2023)
Volume 17: 2 Issues (2022)
Volume 16: 4 Issues (2021)
Volume 15: 4 Issues (2020)
Volume 14: 4 Issues (2019)
Volume 13: 4 Issues (2018)
Volume 12: 4 Issues (2017)
Volume 11: 4 Issues (2016)
Volume 10: 4 Issues (2015)
Volume 9: 4 Issues (2014)
Volume 8: 4 Issues (2013)
Volume 7: 4 Issues (2012)
Volume 6: 4 Issues (2011)
Volume 5: 4 Issues (2010)
Volume 4: 4 Issues (2009)
Volume 3: 4 Issues (2008)
Volume 2: 4 Issues (2007)
Volume 1: 4 Issues (2006)
View Complete Journal Contents Listing