03-05-2011, 12:13 PM
Submitted by:
Badri Adhikari Md.
Hasan Ansari
Priti Shrestha
Susma Pant
Abstract
“Neurology Diagnosis System” is a web-based expert system for diagnosis of neurologic disorders or the disorders of our nervous system. Health assistants in remote areas can use the system to diagnose neurologic patients in the absence of neurologists. They do not need to be neurology expert to diagnose common neurologic disorders if trained to use this expert system. The project is of immense importance to provide medical care facilities to rural people. The importance is felt because medical experts are not readily available in local hospitals and rural areas. These medical care requirements can be fulfilled to a large extent by the use of expert system. This claim is made based upon the results that the system shows and also upon the feedbacks received from various doctors and experts who observed the system. Concerning the methodology, the basic principle is to encode the knowledge of the neurology domain in the form of rules and representative cases and use this knowledge to solve new cases of patients. This knowledge encoding is essentially the implementation of two artificial reasoning techniques called case-based reasoning and rule-based reasoning and is ultimately used to assist the diagnosis process. Since the software uses artificial reasoning techniques it is an artificial intelligence project that exploits theoretical knowledge in the form of rules and practical experiences in the form of representative cases. To perform the reasoning, input to be provided is the clinical examination data along with medical history. The result of performing rule-base reasoning is a list of probable diseases and the output of case-based reasoning is a list of cases similar to the one being diagnosed. Important of the two is the case-based reasoning component that is like a learning machine that keeps learning as new cases are inserted. The web-based hybrid expert system was built using the Java programming language by implementing the Spring Web MVC framework with MySQL as the database system. Keywords: medical diagnosis system, artificial intelligence, rule-based reasoning, case-based reasoning, hybrid expert system, neurology, quadriplegia, paraplegia, nearest neighbor algorithm.
1 INTRODUCTION
1.1 Background
Our mental capacities are so important to our everyday lives and our sense of self that humankind has given itself the scientific name homo-sapiens; man the wise. The field of artificial intelligence (AI) attempts to understand these capacities better known as intelligent entities. AI being a broad topic combines computer science, physiology and philosophy. One of the large scale applications of the field of AI is the development of expert systems. An expert system is a knowledge-based computer program containing knowledge of human experts in a particular domain. Among the various types of expert systems, this project reports the development of a hybrid expert system which is an integration of rule-based reasoning and case-based reasoning techniques. Rule-based system is used when problem area is narrow and the domain has well-understood theory. To overcome these limitations, case-based reasoning technique is also integrated. This solves new problems based on the solutions of similar past problems rather than merely using rules. These expert systems are extensively used in the various fields of medicine, especially for diagnosis purposes. They are already implemented to help the diagnosis process relating to blood infections, heart problems and kidney disorders. This project concerns the implementation of a system for the neurology domain. Neurology deals with disorders of the nervous system with all categories of disease involving the central, peripheral, and autonomic nervous systems, including their coverings, blood vessels, and all effectors tissue, such as muscle.
1.2 Objectives
Following were the main objectives of the project.
1. To develop a web based hybrid expert system to help the neurology diagnosis process.
2. To review Artificial Intelligence literature in Expert Systems and estimate the Expert System model that fits in the domain of neurology.
1.3 Project Overview
This project is about a system that helps in the diagnosis of neurologic disorders. It uses artificial reasoning techniques to assist the diagnosis process. A person will require at least some familiarity with the neurology domain to use the system. This is because the system is built around clinical examination terms and medical jargons that common people are not familiar with. The system implements two major reasoning techniques that human beings use, to assist the diagnosis process. One part of the system, called the rule-based component, is embodied with the diagnosis knowledge that guides the user to find probable diseases in step-by-step process where options are provided at each step. The other part, called the case-based component is like a learning machine that makes better predictions every time it encounters a new case. The two components operate independently and may be used one at a time. Reasoning performed by recalling the theories that we have learnt is called rule-based reasoning and that by remembering similar cases or scenarios is case-based reasoning. Rule-based reasoning is implemented in the system by encoding the domain knowledge into the database and case-based reasoning is implemented such that the system keeps learning as it encounters new cases. Neurology is a medical specialty that deals with disorders of the nervous system. Interested users like health assistants can use the website as a helpful tool to diagnose patients or even themselves. The web application was built by encoding rules of the neurology domain and by developing a framework to learn from the cases of the patients. Integrating the techniques of rule-based reasoning and case-based reasoning a hybrid system was constructed.
Download full report
http://neurologydiagnosissystem.googleco...%20v13.pdf
