About the Book

Intelligent techniques based diagnostic systems have played a vital role in medicine. From statistical techniques to data mining algorithms to neural networks, all these have been widely deployed on medical records for predicting the sickness. Due to increasing vagueness and complexities in health examination data, deriving intelligible information becomes a major challenge for physicians. This challenge could lead to imprecise assessment of the disease and would further direct inaccurate treatment to patients. Therefore to avoid these uncertainties in interpretation of multifaceted data up to a feasible extent, medical professionals employ intelligent techniques based prediction models. Like for other health complications, intelligent techniques have also shown significant performance in the diagnosis and classification of liver disease.

Liver is the largest internal organ in a human body which performs numerous metabolic functions. It filters blood, aids in digestion of fats, makes proteins for blood clotting and most importantly detoxifies harmful chemicals. Liver has a vital importance to life but improper functioning of it may cause serious health consequences. Liver disease is usually caused by inherited disorders, contaminated food, damaged hepatocytes infected with viruses, bacteria or fungi, excessive fat accumulation, and excessive consumption of alcohol or drugs. It is a serious area of concern in the universal set of medicine and is becoming the leading cause of death in India, as well as in other countries around the globe. Ability of liver to resist early detection, as it functions normally even when partially damaged, makes the disease even more alarming because by then it might have suffered eternal damage. This indicates that an early diagnosis of liver disease is a necessity so that in time treatment can be initiated. During diagnosis, analyzing complex medical records of patients may lead to erroneous evaluation and may stretch the decision time of doctors. To overcome these obstacles, computational models are developed using a variety of intelligent techniques which eventually assists the physicians in the diagnostic process.

The aim of this thesis is to build intelligent techniques based computational models for the diagnosis and classification of liver disease, and to analyze their performance using statistical parameters. The motivation behind this work is to assist physicians in liver disease evaluation process, to overcome liver biopsy up to a possible extent, to efficiently analyze complex and ambiguous health examination data of patients, and to reduce the cost, time and effort needed. The intelligent models are developed for identifying liver disease, predicting degree of liver damage, classifying primary biliary cirrhosis, diagnosing hepatitis disease, and classifying alcoholic liver damage, primary hepatoma, liver cirrhosis and cholelithiasis. These models are built using various categories of intelligent algorithms include dimensionality reduction methods, clustering techniques, instance based methods, decision trees, rule system methods, and ensemble learning approaches. Experimental results prove the credibility of proposed computational models in performing non-invasive automatic diagnosis and classification of liver disease. It is observed that the models have the capability of assisting physicians in examining patient liver health by making most efficient use of limited resources. For future perspective, these intelligent systems can also be deployed for predicting human disease other than liver.