INTRODUCTION:

Diabetes mellitus is a multisystem disease related to abnormal insulin production, impaired insulin utilization, or both Diabetes mellitus is a serious health problem throughout the world¹. Diabetes Mellitus is a silent disease and is now recognized as one of the fastest growing threat to public health in almost all countries of the World. It is also called the disease of prosperities. Prevention is better than cure and is less expensive.^2,3

Excess sugar consumption will affect the body's resistance with predicted dangers such as diabetes.⁴ Diabetes Mellitus (DM) is a disease characterized by increased levels of glucose in the blood caused by impaired secretion or action of insulin⁵. Diabetes mellitus is a form of immune inhibition which is actually a very important system for the human body^6,7.

Diabetes mellitus can trigger various complications of dangerous diseases and require long-term treatment. However, many people do not have a high awareness of the dangers of this disease and also do not have basic knowledge about this disease and experience limited time so they rarely consult a doctor⁸. Even the World Health Organization (WHO) reports that 60% of deaths of all ages in the world are non-communicable diseases (NCDs). Diabetes mellitus is the sixth leading cause of death. Around 1.3 million people die from diabetes and 4 percent die before the age of 70⁹.

In Indonesia alone Epidemiologically, it is estimated that the prevalence of Diabetes Mellitus (DM) in Indonesia reaches 21.3 million people. Based on research results, it is estimated that the number of people with diabetes in the world will reach 300 million people¹⁰. Not only that, complications caused by DM can be in the form of macrovascular complications such as cardiovascular disorders, and microvascular complications which can be retinopathy, nephropathy, or neuropathy¹¹. Patients who experience complications in type 2 DM have the potential to get prescriptions with various drugs¹². Prescriptions given by doctors are often found to have DRP events, one of which is drug interactions. Drug interactions are one of the main problems for patients receiving polypharmacy therapy¹³ . Several previous studies have shown that according to Marquita et al, 2014 cited by Eni Faristin (2017)¹⁴, the possibility of drug interactions increases 2.5 times for each drug added to the patient's prescription and on the side effects of individuals with diabetes mellitus including more susceptible to drug interactions. A drug interaction is one of the drug-related problems identified as an event or state of drug therapy that can affect a patient's clinical outcome.

Drug interactions occur when the pharmacokinetics or pharmacodynamics of drugs in the body are affected by one or more interacting substances¹⁵. During the Covid-19 pandemic, people are afraid or worried about traveling out of the house or to the hospital, therefore the expert system makes it easier for people to detect DM symptoms^16,17,18. An expert system is a computer program that imitates thought processes and expert knowledge to solve a specific problem¹⁹. The implementation of expert systems is widely used for commercial purposes because expert systems are seen as a way of storing expert knowledge in a particular field into computer programs in such a way that they can make decisions and reason intelligently²⁰. One implementation that can be applied is in the fields of pharmacology and therapy²¹.

Drug interactions are considered clinically important if they result in increased toxicity and/or reduced effectiveness of the interacting drugs, especially when it comes to drugs with a narrow safety margin (low therapeutic index)²².

Based on the above problems which tend to have a high increase, as well as problems related to polypharmacy that are difficult to avoid, it is necessary to monitor drug therapy in Diabetes Mellitus patients, by conducting studies related to drug interactions with DM patients using digital instruments in the form of expert systems to minimize the possibility of side effects.

MATERIALS AND METHODS:

Based on the level of investigation, this research is literary or uses a systematic literature review or also called descriptive analysis based on research data on the use of expert systems as a source of information in the detection of Drug Interactions Against Diabetes Mellitus Patient Therapy through a literature search from Google Scholar.

The search criteria used are documents in the form of articles during the 2010-2021 period. The year 2010 is used as the initial reference as the minimum year in writing this journal. All selected articles contain research on the theme of diabetes mellitus and its treatment interactions. While the special keywords used are expert systems, diabetes mellitus, treatment therapy, and drug interactions.

Based on the contents of the articles known through abstract searches and discussions, the distribution of research is grouped according to predetermined criteria. The main content of each article is explained at a glance with a descriptive analysis model using paragraphs. The following is the flow of the systematic literature review used in this research:

Figure 1. Flow of systematic literature review

Several stages are starting from planning, then proceeding with looking for articles and then filtering articles that meet the criteria, after that the review and analysis process, and the last is reporting or writing the research results that have been obtained. So, this study tries to map the distribution of research on the use of expert systems as a source of information in the detection of drug interactions with therapy for diabetes mellitus patients and then offers several frameworks.

RESULT AND DISCUSSION:

Components of expert system:

Expert System is one of the most used artificial intelligence applications. It is a set of programs for manipulating knowledge to solve problems in specific domains that require human expertise. Expert systems are also called knowledge-based systems. It is a computer curriculum whose functions and rules are used for data-related reasons²³.

The expert system created is a system based on rules where the program is stored in the form of rules as a problem-solving procedure²⁴. Expert systems can be applied to solve various problems. Generally, the speed in solving problems in an expert system is relatively faster than that of a human expert. This has been proven in several world-famous expert systems²⁵.

According to (Ramanda, 2015) an expert system is composed of two main parts, namely²⁶:

a. Development environment, The expert system development environment is used to incorporate expert knowledge into the expert system environment.

b. Environment Consultation, The consulting environment is used by users who are not experts in acquiring knowledge.

Figure 3. Expert system components

The components of the expert system consist of: 1) The user interface functions as a medium for input of knowledge into the knowledge base and communicates with the user. The interface receives information from the user and converts it into a form that can be accepted by the system. In addition, the interface receives information from the system and presents it in a form that can be understood by the user; 2) Knowledge Base contains all the facts, ideas, relationships, and interactions of a particular domain; 3) The inference engine is in charge of analyzing knowledge and conclusions based on the knowledge base²⁷. One of the important parts in making the application is the design stage.

The purpose of developing an expert system is not actually to replace the human role but to substitute human knowledge into a system form so that it can be used by many people²⁸.

Utilization of Expert Systems in the Health Sector:

The application of technology in the medical field has become commonplace in assisting doctors in diagnosing diseases²⁹. This technology is very helpful in all aspects such as data archiving and information media. One of the current technological trends in expert systems. An expert system is a computer-based system that uses knowledge, facts, and reasoning techniques to solve problems that usually can only be solved by an expert in a particular field³⁰. Today, Information technology fits in all social contexts, and all sciences and is growing rapidly. Information technology is increasingly being used in modern medical practice, health care management, and knowledge of medical professionals ³¹. One of the parts related to artificial intelligence is the decision support system which plays a major role in this field. One of the areas that the decision support system enters is the health sector³².

A clinical decision support system is an interactive computer program designed to assist doctors and other healthcare professionals in deciding³³. Expert systems are very helpful for decision making, where this expert system can collect and store knowledge from a person or several experts in a knowledge base and use a reasoning system that resembles an expert in solving problems³⁴. Some expert systems use open source platforms for implementation. The ES implemented in this is an open-source, web-based system with Google App Engine and cloud computing techniques to manage diabetes and provide patient advice/treatment³⁴.

In one of the studies, the use of expert systems in the health sector is one of them to determine the type of drug for patients with a disease. The test value in the form of a percentage value from the calculation results based on the Certainty Factor formula which produces an accuracy percentage value of 98%³⁵. Some examples of other studies that use the Expert System Design method, one of which is Diabetes Mellitus Diagnosis Using the Web-Based Forward Chaining Method. This method provides space for experts to give value to the knowledge they express. Design and build a clinical pathway application for obstetrical disease in pregnant women using the Naive Bayes algorithm^36,3⁷.

Use of Expert System for Drug Interaction Detection:

A drug interaction is defined as an increase or decrease in the medical diagnostic or therapeutic effect of a particular drug caused by another substance³⁵, which may be another drug, plant, or dietary supplement. Mechanisms of drug interactions can be divided into two categories: (1) pharmacokinetic interactions, which affect the absorption, distribution, metabolism, or excretion of drugs (ADME rules) and thereby cause an increase or decrease in plasma drug levels; and pharmacodynamic interactions, which alter the pharmacological efficacy of a drug while the plasma level of the drug remains unchanged³⁶.

Figure 4. Flow of expert system for drug interaction detection

The existence of interactions with most drugs can be a source of information to find out contraindications, where contraindications are situations where medical products should not be given for safety reasons³⁵. This raises the question of whether it is best to avoid the potential risks of contraindicated drugs and leave the disease untreated or accept the risks of maintaining it³⁶. An expert system in this case can be used to assist in making a diagnosis. An expert system is a system that imitates the expertise of an expert in a field³⁷. This system is made using the Forward Chaining method. In its utilization, the expert system uses a knowledge base that is used, namely by rule-based reasoning. In rule-based reasoning, knowledge is represented by using rules in the form of IF-THEN³⁸.

The knowledge base of the problems identified and analyzed, the next step is to choose a knowledge representation method which will later be used to enter the data obtained in the knowledge acquisition stage³⁴. The representation method used is: the facts obtained from experts, science, research, and their experiences in identifying types of drugs³⁹. The application of an expert system for selecting drugs for systemic diseases aims to assist doctors and pharmacy users in choosing drugs that are appropriate to the patient's systemic condition and obtain information about the drugs to be prescribed through a reasoning process for the inputted patient's condition³⁶. The process of selecting drugs in this expert system is based on the results of the diagnosis as well as the presence or absence of comorbidities and or drugs that are being consumed by the patient. In this process, the system will provide a list of facts that have been stored in the system in the form of a knowledge base³⁹.

In one of the studies, Silva (2018) succeeded in developing an Intermed application as detection of drug interactions through a specialist system. The support provided allows rapid manipulation of large amounts of data, standardization of terminology, and drug relationships. offered to users two features: Items and Quest⁴⁰. The Items function allows users to view a list of registered drugs and create new drug registrations. The search function for drug interactions will be performed with decision-support from rule-based machine learning and also based on health guide recommendations⁴¹. The use of an expert system to detect drug interactions in diabetes therapy was also developed in several studies using a database consisting of 4 tables to store the names of antidiabetic drugs and drugs that contradict them, then identify drug interactions between the two drugs, pharmaceutical groups, and interactions between drugs. It can be expressed by simple models and fast programming languages such as C#.NET can also design decision support systems with high-yield implementation³¹. Several studies have also concluded that engineered expert systems can be used for further detection of drug interactions. This expert system needs to be developed again in the form of portable applications or android-based applications so that the community's reach for this system can be even wider⁴².

In addition, the advantages possessed by expert systems are that this method will work well when the problem begins with collecting or unifying information and then looking for what conclusions can be drawn from that information, information from only a small amount of data⁴². Even so, expert systems have the disadvantage that there is probably no way to identify where some facts are more important than other facts⁴³.

CONCLUSION:

Based on the results and discussions that have been obtained, it can be concluded that the expert system is very useful for detecting disease and is also useful for checking drug interactions with treatment therapy. The expert system itself also has various features and advantages as well as weaknesses. And in its utilization, expert systems usually use several methods, one of which is the Forward Chaining method Algorithm Naïve Bary

CONFLICT OF INTEREST:

The authors have no conflicts of interest regarding this investigation.

REFERENCES:

1. Kumar, P., Mangalathil, TX., Choudhary, V. An experimental Study to Assess the Effectiveness of Structured Teaching Programme on Knowledge Regarding the Management of Diabetes Mellitus among G.N.M. Students in selected Nursing School at Sikar, Rajasthan. Asian J.Nur.Edu & Research. 2014; 4(3): 304-306.

2. Choudhary, VS., Chaudhary,G. A Descriptive Study to Assess the Knowledge Regarding Diabetes Mellitus, Its Risk Factors and Complication among the Rural Community Sadiq, Faridkot (Punjab). Asian J. Nur. Edu. and Research. 2015; 5(2): 251-253.

3. Suzanne C. Smeltzer, Brenda G. Bare. Text book of medical surgical nursing. 10^th edition Philadelphia 2004; 1194-1195. Latin cott.

4. Utami, YP., Triayudi, A., Handayani, ETE. The Diabetes Mellitus (DM) Detection Expert System uses the Android-based Forward chaining and Certainty factor Method. JTIK Journal. 2020; 5(1): 49-55.

5. Takewade, RP. Medical Expert Systems For Diabetes Diagnosis : A Survey. Research Gate. 2014; 4(11): 647–652.

6. Umar, R., Mariana, A.R. and Purnamasari, O. Design of an Expert System for Diagnosing Diabetes Mellitus Using Web-Based Forward Chaining Method. Global Sciences Journal. 2017; 7(1).

7. Wardani, A.K., Santoso, N. and Asmara, R.A. Expert System for Diagnosing Diabetes Mellitus. Journal of Polynema Informatics. 2014; 1(1): 65-69.

8. American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2010; 33: 562.

9. Asemi, A., Ko, A., and Nowkarizi, M. Intelligent Libraries: A Review on Expert Systems, Artificial Intelligence, And Robots. Hi Tech Libraries. 2020; 39(2): 412–434. https://Doi.Org/10.1108/Lht- 02-2020-0038.

10. Ashari, and Muniar, AY Application of Expert System To Diagnose Digestive Diseases With Natural Treatment. National Seminar on Science and Technology. 2016; 2407– 1846.

11. Cahyaningsih, I., and Wicaksono, WA. Risk Assessment of Drug Interactions in Patients with Type 2 Diabetes Mellitus. Indonesian Journal of Clinical Pharmacy, 2020; 9(1): 9. https://Doi.Org/10.15416/Ijcp.2020.9.1.9

12. Dahria, M. Expert System Development in Building an Application. Scientific Journal. 2011; 10(3): 199–205.

13. Darlington, KW Designing for Explanation In Health Care Applications Of Expert Systems. Sage Open. 2011; 1(1): 1–9. https://Doi.Org/10.1177/2158244011408618.

14. Doherty, JA, Reichley, RM, Noirot, LA, Resetar, E., Hodge, MR, Sutter, RD, Dunagan, WC, & Bailey, TC Monitoring Pharmacy Expert System Performance Using Statistical Process Control Methodology. In Amia ... Annual Symposium Proceedings / Amia Symposium. Amia Symposium. 2003; 205–209.

15. Suwarso, G.A.F., Budhi, G.S. and Dewi, L.P. Expert System for Pediatric Disease Using Forward Chaining Method. Infra Journal. 2015; 3(2): 18-24.

16. Alkaff, M., Khatimi, H., Sari, Y., Darmawan, P. and Primananda, R. An Android-Based Expert System to Detect Types of ADHD Behavior in Children. Journal of Information Technology and Computer Science. 2019; 6(2): 135-140.

17. Pradana, M.G., Pamekas, B.W. and Kusrini, K. Design of an Expert System for Diagnosing Diabetes Mellitus Using Certainty factor Method Design Expert System for Diagnosing Diabetes Mellitus Using Certainty factor Method. Creative Communication and Innovative Technology Journal. 2018; 11(2): 182-191.

18. Hakim, RR Al, Rusdi, E., and Setiawan, MA. Android Based Expert System Application for Diagnosing Covid-19 Disease: Cases Study of Banyumas Regency. Journal Of Intelligent Computing and Health Informatics. 2020; 1(2): 26. https://Doi.Org/10.26714/Jichi.V1i2.5958

19. Hayati, B., Ariyani, H., and Ruslinawati. Potential Drug Interactions in Prescribing Type 2 Diabetes Mellitus Patients In Pharmacy X (Potential Drug Interactions in Prescribing Type 2 Diabetes Mellitus Patients In Pharmacy X). Journal Of Current Pharmaceutical Sciences. 2020; 3(2): 249–255.

20. Ilyas, I. Disease Diagnosis Expert System and Drug Relationality Amor Pharmacy. 2016. http://Repository.Uin- Alauddin.Ac.Id/2907/(2016).

21. Khanan, Kusumadewi, S., and Ruspita, I. Application of Backward Chaining Inference Method in Backward Chaining Method Implementation In Drugs Expert System. Jiko (Journal of Informatics and Computers). 2020; 3(2): 92–100. https://Doi.Org/10.33387/Jiko

22. Liu, K., Wu, Q., & Wang, PApplication of Expert System for Diagnosing Eye Diseases in Humans With Fitriawati Forward Chaining Method. Resources Science. 2009; 37(1): 355– 360.

23. Marlinda, L., Widiyawati, Widiastuti, R., and Indrarti, W. Expert System For Monitoring Elderly Health Using The Certainty Factor Method. Synchronous: Journal of Informatics Engineering Research. 2020; 5(1): 72. https://Doi.Org/10.33395/Sinkron.V5i1.1065

24. May, M., and Schindler, C. Clinically and Pharmacologically Relevant Interactions of Antidiabetic Drugs. Therapeutic Advances In Endocrinology And Metabolism. 2016; 7(2): 69–83. Therapeutic Advances in Endocrinology and Metabolism. https://Doi.Org/10.1177/2042018816638050

25. Nafisa, S. Implementation of Expert Systems in Pharmacology and Therapy to Support Web-Based Decision Making. 2003.

26. Nasser, A. Medical Expert Systems Survey To Cite This Version: Hal Id : Hal-01610722. International Journal of Engineering And Information Systems (Ijeais). 2017; 1(7): 218–224.

27. Nohria, R. Medical Expert System- A Comprehensive Review. International Journal of Computer Applications. 2015; 130(7): 44–50.

28. Nurhayati. Antihypertensive Drug Selection Expert System and Drug or Food Interaction. Antihypertensive Drug Selection Expert System and Drug or Food Interaction. 2016; 6(1): 7.

29. Oguoma, SI, Uka, KK, Chukwu, CA, and Nwaoha, EC. An Expert System for Diagnosis and Treatment of Mental Ailment. Oalib. 2020; 7(4): 1–22. Oalib. https://Doi.Org/10.4236/Oalib.1106166

30. Pradana, MG, Pamekas, BW, and Kusrini, K. Designing an Expert System for Diagnosing Diabetes Mellitus Using Certainty Factor Method Design Expert System for Diagnosing Diabetes Mellitus Using Certainty Factor Method. Citi Journal. 2018; 11(2): 182–191. Citi Journal. https://Doi.Org/10.33050/Ccit.V11i2.586

31. Rahmawaty, A., and Hidayah, PH Relationship of Drug Related Problems (Drps) Category of Drug Interaction on Drug Use in Type 2 Diabetes Mellitus Patients. Journal Of Pharmacy Stikes Cendekia Utama Kudus. 2020; 4(1): 80–88.

32. Ramanda, K. Application of Expert System To Diagnose Diseases In Pregnancy. Journal of Pilar Nusa Mandiri. 2015; 11(2): 179–185. Pilar Nusa Mandiri. https://Doi.Org/10.33480/Pilar.V11i2.151

33. Rasoolimoghadam M., Safdari R., Ghazisaeidi M., Maharanitehrani MR, Tahmasebiyan, S. Designing Decision Support System To Detect Drug Interactions Type 2 Diabetes. Acta Informatica Medica. 2015; 23(6):336338. Acta Informatica Medica. https://Doi.Org/10.5455/Aim.2015.23.336- 338

34. Salim I., Olowosulu AK, Abdulsamad A., Gwarzo MS, Khalid GM, Ahmad NT, Eichie FE, and Kurfi FS. Application Of Sedem Expert System in The Development of Novel Directly Compressible Co-Processed Excipients Via Co-Processing. Future Journal of Pharmaceutical Sciences. 2021; 7(1). https://Doi.Org/10.1186/S43094-0210025Z

35. Septiaji RY, Suria O. Drug Selection Expert System on Self-Medication Using Bayes Theorem Expert System Drug Selection on Self-Medication Using Bayes Theorem. Journal of Multimedia and Artificial Intelligence. 2017; 1(2): 1–8.

36. Silva, E., and Alves, D. Intermed: A Mobile-Based Expert System For Drug Interactions Analysis Intermed : A Mobile-Based Expert System for Drug Interactions Analysis. In Research Gate. 2011; 305-309.

37. Suhendro. Utilization of Expert System Applications for Diagnosis of Lung Disease. Journal of Informatics. 2018; 11(2): 78–86.

38. Tobin, MJ Asthma, Airway Biology, And Nasal Disorders in Ajrccm. American Journal of Respiratory and Critical Care Medicine. 2004; 169(2): 265–276. Respiratory And Critical Care Medicine. https://Doi.Org/10.1164/Rccm.2312011

39. Undata R., Tahun PM, Nurlaelah I., Mukaddas A., Faustine. Study of Drug Interaction in Treatment of Diabetes Mellitus with Hypertension in Outpatient Installation in Undata Hospital Palu Period March-June 2014 200. Galenika Journal of Pharmacy. 2015; 1: 35 – 41.

40. Masykur F. Implementation of Disease Diagnosis Expert System Diabetes Mellitus Using Web-Based Fuzzy Logic Method. Thesis. 2012. Semarang: Diponegoro University.

Received on 19.03.2022 Modified on 09.05.2022

Research J. Pharm. and Tech 2023; 16(1):328-332.

DOI: 10.52711/0974-360X.2023.00058