Certainty Factor Method for an Expert System for Orthopedic Disease Diagnosis

Health is an important thing in human survival, including bone disease. Some people think that bones are passive, dead tissue, but they are not. Bones and skeleton are a very important part of orthopedics. Bones are not only a framework that strengthens the body but are also part of the structure of joints, as protection for the body, where the ends of muscles are attached. One of the diseases that is widely felt in society is bone disease. Bone disease is a condition that damages the skeleton and makes bones weak and susceptible to fractures. The high rate of diseases that attack the bones is caused by the conditions and behavior of society, such as stress, lack of exercise, wrong diet. Another cause is also ignorance and lack of knowledge about bone disease itself. By using an expert system, patients can save time going to the hospital, because the system replaces experts in their field, especially chiropractors. The method that will be used in this research is the Certainty Factor Method. The results obtained later are diagnostic results based on the symptoms entered and issued in the form of decisions and percentages. The diagnosis results identified Cervical Spondylosis (Nerve Pain) with a confidence level of 89.60%. It is hoped that this expert system will make it easier for patients to diagnose bone disease and carry out initial treatment in treating the disease.


Introduction
Health can be described by physical conditions and mental functions that can work well.Health is important in human survival.A healthy body can make it easier to carry out various kinds of activities without problems.With good conditions, daily life will be calmer and productivity will increase [1].
Health is an important thing in human survival, including bone disease.Some people think that bones are passive, dead tissue, but they are not.Bone is living tissue and grows, and continuously breaks down, reforms and repairs its tissue [2].
Bones and skeleton are a very important part of orthopedics.Bones are not only a framework that strengthens the body but are also part of the structure of the joints, as protection for the body, a place for attachment of the ends of muscles which are attached to bones with constant or stable movement during contraction (origo), the ends of muscles which are attached to bones with steady movement change position during contraction (insertion) and the muscles that move the body frame [3].
One of the diseases that is widely felt in society is bone disease.Bone disease is a condition that damages the skeleton and makes bones weak and susceptible to fractures.The high rate of diseases that attack the bones is caused by the conditions and behavior of society, such as stress, lack of exercise, wrong diet.Another cause is also ignorance and lack of knowledge about bone disease itself [4].
From this problem, as the development of technology and information systems grows rapidly, computers are used and utilized as a medium to provide information and improve the performance of human knowledge.One area that can be utilized in this case is the Expert System.An expert system is a computer system that can imitate the abilities of an expert.An expert is someone who has special skills and can solve certain problems that cannot be solved in general by lay people.Expert systems can combine inference rules with a knowledge base that comes from one or more experts in a particular field [5][6].
One method that can be used to overcome uncertainty is the Certainty Factor method.Certainty Factor is a method that defines a measure of certainty regarding facts or rules to describe an expert's confidence in the problem at hand.Certainty Factor shows a measure of certainty regarding a fact or rule [ 7][8] [9].
Research on expert systems in various fields has been widely carried out.The following are several references used as references in this research.
Ratniasih conducted research on an Expert System for Diagnosing Meningitis Using the Naïve Bayes Method.Meningitis is a deadly disease for sufferers.This disease is caused by bacteria, viruses or fungi.Lack of information about the symptoms of meningitis and the lack of information about this disease are the main causes of the high death rate among the community.So we need a system that can provide information about meningitis.Therefore, there is a need for an expert system to help the public obtain information about the symptoms of meningitis and solutions [10] Agustino also conducted research on expert systems related to the world of health, namely for diagnosing kidney disease.The development of computer software technology, which has developed rapidly and is followed by the development of artificial intelligence technology, allows an expert system to run on websites using PHP applications.This expert system is expected to be able to provide information on all matters related to kidney disease problems quickly and efficiently in a reciprocal manner for both the user and the system but still optimally [11].
Hidayat, et al diagnosed diseases caused by smoking using the implementation of an expert system.Cigarettes are objects that have various sources of disease if we smoke them directly or inhale the smoke.In a cigarette there are more than 4000 chemicals, of which 200 are dangerous to humans, and 40 of which cause cancer.Smoking can cause changes in the function, tissue structure and respiratory tract of the lungs.The system used in writing this final assignment is an expert system for diagnosing diseases caused by smoking and the method used is Forward Chaining [ 12].
From the research described above, this research was carried out using expert system technology.This Expert System aims to be able to help the wider community to diagnose bone symptoms.In an effort to obtain high calculation accuracy so that the final results obtained are optimal, the Certainty Factor method is used using symptom weights selected by the user.

Research methodology
In preparing research, it is necessary to have a framework structure with clear stages, this framework is used to solve a problem that is being discussed by the researcher, the framework is in Figure 1.The process of calculating the level of confidence begins by decomposing a rule that contains several symptoms into a number of rules that only involve one symptom.After that, each new rule will be assessed using a specific formula to calculate the Certainty Factor (CF) value.

𝐶𝐹(𝐻, 𝐸) = 𝐶𝐹 (𝑢𝑠𝑒𝑟) * 𝐶𝐹 (𝑟𝑢𝑙𝑒)
Among the conditions that occur, there are several conditions that have the same result (in different rules), we need to collect or combine the Certainty Factor (CF) values from each existing condition to get the overall CF value using the following equation: If both CF > 0 , then the formula is: The diagnosis results provided by the expert system are the percentage of possible diseases.Determining the disease to be used as a diagnosis is based on the largest percentage among all the diseases evaluated.This disease percentage is generated from calculating the Certainty Factor value based on the symptoms selected by the user.

. Data analysis
The data analysis stage is the most important stage in developing a system, because at this stage an evaluation will be carried out to identify existing problems, system design and the steps needed for the desired design to arrive at the expected analysis.The data variables obtained are a collection of types of disease, disease symptoms and solutions or ways of treating them obtained from various sources such as the results of interviews with doctors at Semen Padang Hospital as well as journals and the internet.The following data will be used as research objects: 1. Symptom Data The symptom data used in the expert system for diagnosing bone disease includes 26 symptoms.The symptoms of bone disease can be seen in Table 1  physiotherapy with heat, ultrasonic waves, or electrical stimulation ( electrical therapy ) can help overcome back pain.After the pain is resolved, physiotherapy can be continued with special movement exercises to increase the strength and flexibility of the back muscles.d.Acupuncture, this alternative therapy is believed to help reduce lower back pain.The results will be better if acupuncture is combined with physiotherapy and medication.e. Spinal surgery, in serious cases, such as spinal structural abnormalities or pinched nerves that do not improve with treatment, surgery can be performed to correct abnormalities in the nerves and spine.
5 Osteoarthri tis a.To relieve pain and inflammation the doctor will give medicines, such as paracetamol, non-steroidal antiinflammatory drugs such as ibuprofen, capsaicin cream, injections of corticosteroid drugs.b.Apart from giving the medicines above, osteoarthritis can also be treated with physiotherapy to strengthen the muscles around the joints, this method can also increase the flexibility of the joints and muscles, and reduce pain and surgery can be done to repair or replace damaged joints.sufferers can move more easily.
Examples are total hip replacement in hip osteoarthritis and total knee replacement in knee osteoarthritis.c.Apart from treatment provided by a doctor, osteoarthritis sufferers are advised to adopt a healthy lifestyle, such as regular exercise and losing weight.

Relationship between Disease and Symptoms
The following is the relationship between disease and symptoms which can be seen in table 4 below: Table 4. Relationship between disease and symptoms

Process Analysis
The process stage uses the certainty factor method calculation where the symptoms of bone disease have an expert CF value.Diseases, symptoms and expert CF values used in this study can be seen in table 5 below:  The expert CF value for each symptom is different, but there are also similar values .The value for each symptom describes the influence of that symptom on bone disease.The greater the symptom value, it can be concluded that the symptom is the dominant symptom of disease in people with bone disease.Thus, from tracing the rules and calculating the certainty factor, it can be said that the patient was identified as having Cervical Spondylosis (Pain in the Neck Bone) with a confidence level of 89.60%.

Conclusion _
After building a web-based expert system at Semen Padang Hospital, users/patients can diagnose bone diseases quickly and easily so that users/patients can minimize the time they have to come to the hospital.Bone disease diagnosis data can be stored properly and safely.After designing an expert system for diagnosing bone disease based on the symptoms felt by the patient, we can get a solution for initial treatment in treating bone disease.Helping patients diagnose bone disease so that it can be done correctly and accurately.

Table 1 .
Symptom Data The number of diseases processed in the bone disease diagnosis expert system is five types of disease, as in

Table 2 Table 2 .
Data on Types of Disease

Table 3 Table 3 .
Solution Data

Table 5 .
Diseases, Symptoms and Expert CF Values