clinical diagnosis of enteric fever by computer expert systems

1. Introduction

Clinical diagnosis forms the climax of the decision making process in clinical care. And clinical care is the provision of what is necessary for a person's health and well-being by a clinician. So clinical diag-nosis is a part of the entire clinical care process. It is the pivotal cognitive activity of a practicing physician [1]. In clinical (or medical) diagnosis, a physician uses several sources of data and puts the pieces of the puzzle together to make a diagnostic impression. The initial diagnostic impression can be a broad term de-

Nig J. Biomed Engrg, 2020, 13, 21 - 32 21

*Corresponding author email: [email protected]

Clinical Diagnosis of Enteric Fever by Computer Expert Systems

1Nkuma-Udah K. I.*, 2Chukwudebe G.A., 3Ekwonwune E., 1Ejeta K.O., 4Onwodi G.O. and

1Ndubuka G.I.

1. Department of Biomedical Technology, School of Health Technology, Federal University of Technology, Owerri, Nigeria

2. Department of Electrical / Electronic Engineering, School of Engineering and Engineering Technology, Federal University of Technology, Owerri, Nigeria

3. Department of Computer Science, Faculty of Science, Imo State University, Owerri, Nigeria 4. Department of Computer Science, Faculty of Science, National Open University of Nigeria, Abuja, Nigeria

(Received May 20, 2020; Revised June 05, 2020; Accepted September 05, 2020)

Abstract Missing medical data abound in developing countries leading to mis-diagnosis and inappropriate treatment. These conse-quently trigger a chain reaction of repeat diagnosis, re-treatment, missed disease outbreak recognition and eventually high cost of medical care. This is worse when the mis-diagnosis is on an endemic, infectious and preventable disease that is one of the leading causes of death in the developing countries, such as enteric fever. The aim of this research is to de-sign a an Expert System for the diagnosis of Enteric fever called ENTERASE, to enable clinicians to adequately collect medical data relating to enteric fever, process them into information, integrate and interpret the information so as to lead to proper diagnosis and treatment of enteric fever disease. Enteric (tyfoid) fever is chosen for its public health as well as socioeconomic importance in a developing country like Nigeria. Automation of medical diagnosis to advance the course and quality, and eventually improve the overall outcome of the medical care is mostly beneficial to the remote areas of developing countries, where the population is deprived of the facilities of having several medical experts to diagnose diseases. In this design, ENTERASE, the expert system uses the iteration method and employ a rule-based, forward-chaining and the Visual Basic 6.0 programming language. Knowledge base was built by accumulating factual knowledge from literature and medical experts of the enteric fever domain. Knowledge in the design was represented via a produc-tion rule and used as a base for analysis, diagnosis and treatment recommendations. The ENTERASE provides a simple, interactive, graphical user interface with menu. The system uses plain English language to interact with user and so it is expected that its use will eliminate missing data, enhance the diagnosis of enteric fever disease and so reduce the prob-lems that come with missed medical data, mis-diagnosis and their eventual mortality causes. Keywords: Design, Expert systems, Medical Diagnosis, Malaria, Knowledgebase, ENTERASE, VB 6.0

scribing a category of diseases instead of a specific disease or condition. After the initial diagnostic im-pression, the physician obtains follow up tests and procedures to get more data to support or reject the original diagnosis and attempts to narrow it down to a more specific level using the diagnostic procedures [2]. 1.1 Medical Care and Diagnosis

Medical care has hitherto been consumer-focused, where the medical care providers are the experts, family are visitors, and patients are body parts to be

ABC Publications Incorporated

AB

CP

ubl ic

ationsI nco rp

or ate d

Dir ec

to riesof A fricanBio

grap

hy

*

A B C

*

fixed. Today, medical care is patient-centered, where the delivery of care is organized around the needs of the patient. The defining characteristics of patient-centered care is the partnering with patients and families, welcoming - even encouraging - their in-volvement, and personalizing care to preserve pa-tients’ normal routines as much as possible [3]. A typical medical care process goes through the Medical / Clinical Care Process Cycle as shown in Figure 1. From the cycle, the medical (clinical) care process involves: Initial presentation: patient comes to physician for consultation. Data: gathered from patient through interview, ob-servation, instrumentation, monitoring and telemetry by the physician. Information: data is converted to information by its interpretation, filtering, sampling, smoothing and clustering Diagnosis: From the information gathered, diagnosis can be made by inference, model-based reasoning and classification Therapy: With the diagnosis made, treatment can be instituted by planning, predicting effects and antici-pating certain outcome. Outcome: The treatment instituted can be good or bad - prognosis.

The part of the medical life cycle this re-

22 Nkuma-Udah K. I., Chukwudebe G.A., Ekwonwune E., Ejeta K.O., Onwodi G.O. and Ndubuka G.I.

search is interested in, is Diagnosis. Medical diagno-sis as part of the medical care cycle is a cognitive process, where a physician uses several sources of data and puts the pieces of the puzzle together to make a diagnostic impression [1]. The initial diag-nostic impression can be a broad term describing a category of diseases instead of a specific disease or condition. After the initial diagnostic impression, working diagnosis, the physician obtains follow up tests and procedures to get more data to support or reject the original diagnosis and attempts to narrow it down to a more specific level using the diagnostic procedures. A typical diagnostic process is shown in Figure 2.

Patient

Initial

presentation

Physician

Interpret Data Information

Formulate

Plan

Diagnosis Therapy

Prognosis

Outcome

Figure 2.3: The Medical (Clinical) Care Process Cycle Figure 1: The Medical (Clinical) Care Process Cycle

Fig

ure

2:

The D

iagnost

ic P

roce

ss (

So

urc

e: I

mp

rov

ing

Dia

gn

osi

s in

Hea

lth

Car

e b

y T

he

Nat

ion

al

Ac

adem

ic P

ress

)

In a typical Diagnostic Process, a physician usually undertakes processes of Information gather-ing and then Information integration and interpreta-tion, which leads to a working diagnosis used to in-stitute Treatment. Sometimes the working diagnosis is not single but a number of diagnosis called differ-ential diagnosis. As the number of differential diag-nosis increases, the diagnostic (decision making) process, especially at the information gathering and information integration / interpretation stages, in-creases in complexity. With the increasing complexity, the decision making process (diagnosis) becomes so difficult that more clinical information may be needed to confirm a clinically suspected diagnosis [4]. For example, in some developing countries where many febrile dis-eases are endemic, confirming these diseases caus-ing fever may be difficult. And unless the exclusion of other causes of fever on medical history, physical examination and on other laboratory examination are carried out, definitive diagnosis of a single febrile disease will still be a mirage [5]. In this situation, there is the need for an intelligent healthcare infor-matics to be employed to enable very accurate re-sults by way of definitive diagnosis. Now, an intelligent healthcare informatics is used to mean an expert system, which is a computer program used for medical diagnosis. It performs complex data processing similar to evaluation made by a human expert like the medical doctor [6]. The program is able to draw conclusions and make deci-sions, based on knowledge, represented in its data-base. Expert systems currently play important roles in medicine, medical practice and medical or health-care [7]. With the expert systems employed in medi-cal diagnosis, diagnostic results (based on the knowledge-base of many human experts) are ex-pected to be more accurate and therefore more reli-able than those of a single human expert. This in turn is meant to prevent mis-diagnosis with the at-tendant mis-treatment and so reduce the overall cost of medical care. 1.2 The Public Health Concerns of Enteric fever

Many diseases have fever as their major presenting


clinical symptom. In addition, some of these febrile diseases, so called, have many more presenting symptoms almost non-distinguishably in common with each other and are said to constitute the differ-ential diagnoses of each other. The well known feb-rile diseases that are endemic in Nigeria and other developing countries are malaria, enteric (typhoid) fever and dengue fever. The scope of this research is enteric fever. Like malaria, enteric (or typhoid) fever is endemic in Nigeria, a life-threatening systemic in-fection, has high fever as a main symptom and often associated with poor living conditions [8][9][10]. Enteric fever is a global public health problem with an estimated 11-20 million new cases and between 128,000 and 161,000 deaths recorded globally as at 2000 [11][12]. This is especially worse in the devel-oping nations of the world, especially in populations that lack access to safe water and adequate sanita-tion, where it is a significant contributor to morbid-ity and mortality [13][14]. Poor communities and vulnerable groups including children are at highest risk [15]. Enteric fever is caused by the bacterium, Sal-

monella typhi, a Gram-negative, motile, aerobic, nonsporing, intracellular Bacillus also known as Sal-

monella enteric because it grows in the intestines and blood [15]. A similar but often less severe dis-ease, paratyphoid fever, is caused by Salmonella

Paratyphi A, B or C. Transmission of the infection is usually through ingestion of contaminated food or water [16]. Risk factors include poor sanitation and poor hygiene. Those who travel to the developing world are also at risk and only humans can be in-fected. The acute illness is characterized by pro-longed fever, headache, nausea, loss of appetite, and constipation or sometimes diarrhoea. Symptoms are often non-specific and clinically non-distinguishable from other febrile illnesses. The enteric infection has only human type. Infection begins when an infected food or beverage is eaten, contaminated hand enters the mouth or sew-age-contaminated water or shellfish enters the mouth. The life cycle of the salmonella bacteria causing enteric fever occurs only in the human body and mainly in the enteric system of the body, hence

the name enteric fever [17]. When present in the gut (enteron), the patho-genic Salmonella species are engulfed by phagocytic cells of the gut, which then pass them through the mucosa and they enter the host's system through the distal ileum. They then travel from the gut into the lymphatic system [18]. The typhoidal salmonella reproduces using the host’s cellular machinery through the lymphatic system and then to the re-ticulo-endothelial tissues of the liver, spleen, bone marrow, and lymph nodes. After their replication, they break out into the bloodstream to invade the rest of the body. The sal-monella bacteria then infect the gallbladder and re-enters the gastrointestinal tract in the bile. Those bacteria that do not re- infect the host are typically shed in the stool and are then available to infect other hosts. The simplified life cycle of enteric fever bacteria is shown in Figure 3.

1.3 Medical Diagnosis of Enteric Fever As at 2014, approximately 21 million cases and 222 000 typhoid-related deaths occur annually world-wide, especially in the developing countries [13]. Typhoid outbreaks may be large and prolonged. Ty-


phoid incidence varies in place and time between and within countries. Thus, heterogeneity is a critical consideration for vaccination strategy [19]. Typhoid fever can be prevented and can usually be treated with antibiotics. Surgery is usually indicated in cases of intestinal perforation. Therefore proper diagnosis and treatment of enteric fever, a preventable cause of death, is impor-tant for the developing countries like Nigeria and more so for the rural and remote areas where the population are deprived of the facilities of having several medical experts to diagnose this disease. The dearth of several medical experts in the remote areas can be compensated by the employment of computer expert systems, so that in expert systems, the exper-tise of specialists is stored in computers through the use of expert system technology and few rural medi-cal doctors can achieve the same result compared to when there are several medical experts[20]. The present study, ENTERES, is a system designed specifically for the disease of, enteric fe-ver. This disease is chosen for its public health as well as socioeconomic importance in a developing country like Nigeria, which is the scope of the study. The disease is endemic and a life-threatening sys-temic infection; has fever as a main symptom; mos-quito as its vector of transmission; and transmitted via the faeco-oral route [15][21][22][23]. 1.4 Use of the Computer Systems to Diagnose En-

teric Fever

This work is a computer expert system designed spe-cifically to diagnose enteric fever and so called En-

teric fever Expert System (ENTERASE). The expert system is developed using the Microsoft Visual Ba-sic 6.0 (VB 6.0) in Windows platform. Visual Basic is a third-generation event-driven programming language and integrated devel-opment environment (IDE) from Microsoft Incorpo-rated [24]. VB 6.0 is a very good tool for expert sys-tems. It has a user friendly interface: the Graphical User Interface (GUI). It may be executed in three ways: interactively using the GUI interface; interac-tively using a window/menu/mouse interface on windows, or as embedded expert system in which the user provides a main program and control execu-

Figure 3: Li fe cycle of Enteric fever bacteri a

tion of the expert system. The present design em-ployed execution interactively using the GUI inter-face.

2. Methodology

The Rule-based systems methodology is used, to stipulate the step by step procedure undertaken to arrive at the diagnosis and treatment of enteric fever with their possible prognoses: analysis of the present (conventional) system; identification of problems of the current system; analysis of the proposed system; and system design of present system and its feasibil-ity. 2.1 Analysis of the Present System

Conventionally, enteric fever is diagnosed by a medical doctor based on his prior training and

knowledge on the disease. This is purely a manual system and entirely dependent on a single human expert, who has a lot of limitations and disadvan-tages. These limitations include the fact that humans are inconsistent, forgetful, unable to comprehend large amounts of data quickly, can get tired, have limited knowledge and memory [25]. Figure 4 shows a typical data flow diagram of the manual diagnostic process. In the diagram, the information-giving steps include: initial creation of patient folder Create folder, Data collection of pa-


tient (from signs, symptoms and clinical investiga-tion), Information Integrating & Interpretation of collected data to form a diagnosis, Cognition of the working diagnosis and Consultation with the patient for treatment and lastly the final Outcome of the treatment. From the data flow above, it is clear that the outcome of Diagnosis and eventual treatment de-pend solely on steps of those of Data collection and Information Integration and Interpretation. This means that these two steps (Data collection and In-

formation Integration and Interpretation) are the rate-limiting steps of the diagnostic process. Thus any process or event that causes these rate- limiting steps to malfunction, will lead to poor prognosis. The opposite is also true: any process, event or even intervention that causes the two rate-limiting steps to function well will lead to a good prognosis. 2.2 Proposed Solution to the Present System:

ENTERASE

The proposed solution to the present system is a rule based medical expert system for enteric fever diag-nosis using Visual Basic 6.0 as the programming language. The system, called Expert System for the diagnosis of Enteric fever (ENTERASE), employs a forward chaining inference mechanism and has a good graphical user interface interactive system where systems communicate with user in response to a click of a mouse. The ENTERASE expert system Shell, con-sists of the user interface, the explanation system,

the inference engine and the knowledge base editor. The complete architecture of the proposed expert Figure 4: Overall Data flow diagram of the Present System

User Interface

User

ESMD Shell

Explanation System

Infer enc e Engine

Knowledge Base Editor

Knowledge Base

Case Specific Data Working Storage

Figure 3.3: Architecture of the Proposed System Figure 5: Architecture of the Propos ed System

ENTERASE Shell

system is as shown in Figure 5. The inference engine uses problem solving methods that interacts with the user and processes the result from the collection of rules and data in the knowledge base. The system has the capacity to acquire, store, retrieve, communi-cate, process and use knowledge for the purpose of solving problem. 2.3 Analysis of the ENTERASE

Factual knowledge was accumulated for the expert system knowledgebase via three sources: literature, internet and human experts (medical doctors) in the relevant medical specialties. A set of questionnaires was administered to twenty medical doctors, who are experts in the enteric fever domain selected ran-


domly from membership of Abia State Chapter of Nigeria Medical Association over a period of three months in South Eastern Nigeria. During analysis, the possible symptoms, signs and investigation reports of these diseases were coded and then categorized in groups. Each disease coding is provided with three groups of symptoms/signs/investigation results used as deter-mining factors in the diagnosis as follows: 1) Sr = Sign/Symptom/Investigating report is Strongly Required for Diagnosis; 2) Rn = Sign/Symptom/Investigating report is Rele-

vant but Not necessary for diagnosis; 3) Nr = Sign/Symptom/Investigating report is Not

Related/Relevant for diagnosis.

Table 1: Table of Symptoms of Enteric fever

Symptoms Enteric (Typho id) fever

*fever, paroxysmal (Sr)

* headache dull frontal (Sr)

* malaise (stupurous) (Sr)

* retro-orbital pain (Nr)

* cough (Rn)

* painless skin chancre (Nr)

* anorexia (Nr)

* fatigue (Nr)

* arthralgia (Nr)

* constipation (Rn)

* abdominal pa in (Sr)

* weakness (Nr)

* erythematous skin lesions (Nr)

* myalgia (Nr)

* weight loss (Nr)

* shaking chills (Nr)

* nausea and vomiting (Nr)

* haemoptys is (Nr)

* facial oedema (Nr)

* sweating (Nr)

* rash (Nr)

* sore throat (Nr)

* transient urticarial (Nr)

* lymphadenopathy (Nr)

* chest pain (Nr)

* back pain (Nr)

Table 2: Table of Signs of Enteric fever

Signs Enteric (Typhoid) fever *fever, paroxysmal (Sr)

* peritonit is (Sr)

* malaise (Sr)

* jaundice (Rn)

* cough (Rn)

* intest inal perforation (Sr)

* unconsciousness (Nr)

* haemorrhage (Rn)

* fever (Sr)

* epistasis (Nr)

* high fever (Sr)

* weakness (Nr)

* erythematous skin lesions (Nr)

* cyanosis (Rn)

* weight loss (Nr)

* shaking chills (Sr)

* vomit ing (Nr)

* haemoptysis (Nr)

* mild fever (Sr)

* sweating (Rn)

* skin rash (Nr)

* finger clubbing (Rn)

* lymphadenopathy (Nr)

The knowledgebase was used for analysis, diagnosis and recommendations. Knowledge in the

design was represented via a production rule. Table 1, 2 and 3 display respectively the symp-toms, signs and possi-ble clinical investiga-tions for enteric fever as accumulated in ques-tionnaire and confirmed by knowledge from lit-


erature and internet. The three groups of determining factors of Sr, Rn and Nr for signs/symptoms/investigation reports of enteric were used to diag-nose enteric fever from different production rules. Table 4 shows an example of a Production Rule applying the If, then condition for enteric fever using the signs/symptoms/clinical investigation re-sults of enteric fever in Tables 1, 2 and 3. As an ex-ample, to diagnose enteric fever, if sign/symptom of paroxysmal fever is present and Strongly Required (Sr) for diagnosis, and if shaking chills is also Strongly Required (Sr) for diagnosis, both will be connected by AND. So also will fatigue and positive rapid diagnostics test (and all Strongly Required (Sr) signs/symptoms) for diagnosis will be connected by AND. On the other hand, signs/symptoms like myal-gia, arthralgia and headache, which are all Relevant but Not necessary (Rn) for diagnosis will be con-nected by OR. Finally, signs/symptoms like positive stool culture, which is Not Related/Relevant (Nr) for diagnosis is connected by NOT. Then the diagnosis is confirmed to be malaria. Therefore, diagnosis is performed via the designed expert system, based on patient data put into the system. 2.4 High Level Model of the ENTERASE

The generation of a high level model, HLM, in sys-tem design has the main advantage that it makes it easier for the designer(s) to be able to test the prod-uct against what was originally specified. Such mod-els will usually deal with areas like performance, reliability, availability, maintainability, and system safety [38]. The HLM function to explain the archi-tecture that would be used for developing the soft-ware product. The architecture diagram provides an overview of an entire system, identifying the main

Table 3: Table of Investigations of Enteric fever

Clinical Investigations Enteric (Typhoid) fever

* Packed Cell Volume / Haemoglobin concentration (Rn)

* wbc count (Rn)

* Sputum culture (Nr)

* Fluorescent dyes/ultraviolet indicator tests (Nr)

* PCR may detect dengue virus in serum early in the illness. (Nr)

* Blood Culture (Sr)

* Stool Culture (Sr)

* Urine Culture (Sr)

* Tissue Culture (Nr)

* Blood / Serum Assay (Nr)

* Erythrocyte Sedimentation Rate (ESR) (Nr)

* Acid Fast Bacilli staining (Nr)

• Blood smears ( thin/thick or unstained/Giemsa-stained) (Nr)

* Chest X-ray (Nr)

* Widal test (Sr)

* Rapid diagnostic tests (Nr)

* Platelet count (Nr)

* Aspiration (chancre, lymph node or bone marrow) (Nr)

* Lumbar puncture (Nr)

Table 4: Production Rule for Enteric Fever

If there is high fever (Sr)

AND dull frontal headache (Sr)

AND malaise (Sr)

AND abdominal pain (Sr)

AND positive widal test (Rn)

OR cough (Rn)

OR constipation (Rn) OR stool culture isolated S. typhi

NOT shaking chill (Nr)

Then the Disease is Enteric fever

Log In

Home Symptoms

Signs

Clinical

Investigation

Treatment

Regimen Exi t

Patient Information

Figure 6: High Level Model of the ENTERASE

components that would be developed for the product and their interfaces. Therefore, the high level model of the ENTERASE is as shown in Figure 6. 3. Results and Discussions

3.1 Results

ENTERASE is switched on by double-clicking the system’s program icon ‘expert_medical_sys’ from the system’s folder. A title display first announces the system. This is followed by a log in window,

where the user is expected to click proceed to continue. A booting slash panel is used to open into the login win-dow which requires the

user to insert his username and password. Figure 7a shows the system pictoral while Figure 7b is the sys-tem print screen. If these are correctly inserted, the system opens into the main screen, which contains the main menu.

When the system is accessed, the main menu window containing the sub-system options - Home, Patient information, Records, Diagnostic System, Treatment System and Exit – pops out. The Home Panel option is default and forms the Main Menu. Figure 8a shows the Home Panel screen of the EN-


TERASE pictoral, while Figure 8b is the screenprint from the system.

The Patient Information enables the user to key in new patients’ biodata or to retrieve old pa-tients’ information by supplying their card no and clicking the FIND button. Figure 9a represents the Patient Information screen pictorial while Figure 9b is the print screen of the system. The ENTERASE also uses the user interface to display desired result to users. Visual Basic scripts alerts will display re-sult output to the user. Figure 10 represents the re-sult for Patient Information. The Records option en-ables the user to access previously registered pa-tients’ data. Figure 11 represents the result of click-ing the Patient’s Record. The Diagnostic system option enables the user to decide on whether the patient has enteric fe-ver or not. Here, the user has the option of checking

Figure 7a: Login window of the EN-

USER NAME

PASS WORD

LOGI

Expert System for

the Diagnosis of

(ENTERASE)

Figure 7b: Login window of the ENTERAS E

(ENTERASE)

Medical Diagnosis Expert System (MDES)

for Malaria, Enteric and Dengue fever

Main Menu

Pat ient Information

Home

Records Diagnostic

System Treatment

System

Exit

Expert System for the Diagnosis

of Enteric fever (ENTERASE)

Figure 8a: Home Panel (Main Menu) of ENTERASE)

EXPERT SYSTEM FOR THE DIAGNOSIS OF EXPERT SYSTEM FOR THE DIAGNOSIS OF EXPERT SYSTEM FOR THE DIAGNOSIS OF EXPERT SYSTEM FOR THE DIAGNOSIS OF ENTERIC FEVER ENTERIC FEVER ENTERIC FEVER ENTERIC FEVER (ENTERASE )(ENTERASE )(ENTERASE )(ENTERASE )

Figure 8b: Home Pane l (Main Menu) of ENTERASE)

the symptoms, signs and/or clinical investigation


results of patient under consultation. This is done by clicking and selecting in turn, the Symptoms, Signs and Clinical Investigation results buttons. This en-ables the system to make a diagnostic decision of enteric fever or not with respect to the data put into the system. Figure 12 represents the Diagnostic Sys-tem screen, while Figure 13 represents the output for Diagnosis System. The Treatment System is used to recommend therapeutic regimens, from where the medical doctor (a human expert) chooses the best treatment option based on the diagnosis. Figures 14 and 15 represent the Treatment System input and output respectively. At the end of any patient’s session, the user logs out. The Log out then pops out of the EN-TERASE (Figure 16).

FIND Patient Information

Patient Registration

Card No

Surname

Fi rst Name

Address

Marital Status

Next of Kin

Gender

Figure 4.4: Patient Information Panel of the MDES

Occu pation

Phon e No

Department

Pres enting Co mplaint s

I

Doctor’s Remark

Date of Reg is tration

dd/mm/yy

Date of Last vis it

dd/mm/yy

Date of next v isit

dd/mm/yy

Prognos is

C LEA R DELETE UPDATE R EGIST ER

Fig ure 9a: Patient Informatio n Panel of the ENTERASE (pictorial)

Figure 9b: Patient Information Panel of the ENTERASE (printscreen)

Figure 4.12 Output for Patient Information

Registration successful

OK

Patient Registered successfully i

Figure 10: Output for Patient Information of the MALARES Figure 10: Output for Patient Information of the ENTERASE

Figure 11: Patie nt Record Output of the ENTERAS E

Clos e

Patient’s Record

ID Card No Surname Other Name Address Marital Status Sex

14 01 Nweke Patrick 10 Ojike Lane Aba Married Male

16 02 Peter Ugochi 12 Azuka Drive Aba Sin gle Female

17 03 Okoye Mark 29 Market Road Aba Sin gle Male

Clinical Investigations

Diagnosis

Signs

Symptoms

Click on desired operation to diagnose patient ’s d isease

Figure 4.6: Diagnostic System Panel of the MDES Figure 12: Diagnost ic System Panel of the MALARES Figure 12: Diagnostic S ystem Panel of the ENTERAS E

Diagnostic Confirmation

Figure 4.13: Output for Malaria Diagnosis of the MDES

Malaria Confirmed in Patient

Close

Figure 13: Output for Diagnosis of malaria of the MALARES Figure 13: Output for Diagnosis of enteric fever of the ENTERASE

Enteric fever Confirmed

For the system to function as specified, the various data as knowledgebase are stored in facts list. Different rules exist for different signs and symptoms with investigation results. The system consists of about 41 rules. It provides a simple, inter-active, graphical user and menu based interface and stores all the rules as a batch file. So the series of rules can automatically run directly from a batch file following the click of an icon. The EN-TERASE was tested in stages with volunteer test data. For instance, the Patient Information module was tested by keying in a volunteer’s biodata. The results of the system are as shown in Figures 7, 10, 11, 13, 15 and 16.

3.2 Discussion

A number of works have been done on use of expert


systems for diagnosis of febrile illnesses especially for malaria and typhoid fever. Many are done singly for mainly malaria. A handful are done singly for enteric fever. In his work, Anigbogu adopted the Structured System Analysis and Design Method (SSADM) and the Expert System Methodology to develop the system [26]. Although, the system called TURBO PROLOG, was found to be able to diagnose malaria parasites in patients and recommend treat-ment procedures for such patients. Tunmibi et al de-veloped a project, A Rule Based Expert System for Diagnosis of Fever, Tunmibi, employing a rule based expert system for diagnosing fever [27]. Their web based expert system used Visual Basic Dot Net (VB .Net) as the language of its implementation while the rules within the knowledge base were Boolean rules and not fuzzy rules hence; drawing of inference as performed by this system could not have a high degree of human like way of reasoning. 4. Conclusion and Recommendation

The developed system, ENTERASE is a medical expert system for diagnosing enteric fever. It is ge-neric for enteric fever and can be used by all levels of medical doctors because the signs, symptoms and investigation reports of enteric fever are similar uni-versally. ENTERASE is a rule based system that supports forward chaining inference. Using this sys-tem, the user can enter patients’ signs, symptoms and investigation reports. The system will evaluate it and diagnose whether the patient has enterase fever or not. If the patient has enteric fever, for instance, system will give suggestions for treatment. The ENTERASE is developed using the Vis-ual Basic 6.0 Programming language. It has a simple and user friendly graphical user interface and so does not need intensive training to be used. The sys-tem was evaluated, tested by using classical test cases and checked if the systems result is in agree-ment with the doctors’ diagnoses. The ENTERASE is not meant to replace medical doctors but to assist them in the quality ser-vice they render to humanity. It is also invaluable where younger inexperienced doctors are practicing,

Dengue fever

Treatment

Enteric fever

Malaria

Please click on your diagnosis to get the treatment options

Figure 4.7: Treatment System of the MDES Figure 14: Treatment System of the MALA RES

Diagnosis: No Malaria

Diagnosis: Malaria Confirmed

Diagnosis: Confirmed Enteric fever

Diagnosis: No Enteric fever

Figure 14: Treatment System of the ENTERAS E

# Ant ibiotic: the only effect iv e treatment for ty phoid fever, commonly: * Cip roflo xacin tablet (500mg twice da ily ); * Ceftriaxo ne in ject io n (1 g daily) for h osp italized patients ; * Levoflo xacin tablet 500mg daily; * Chloramph enicol 500 mg tablet four times daily; * Amoxicill in e 25mg/kg four t imes daily; * Trimethoprime-Sulfamethoxazo le 320/1 600 o r 10mg /kg twice dly. # Sy mptomat ic t reatmen t in clude the use of: * drinking fluids; to prevent deh ydration and d iarrhea; * eat ing a healthy diet etc. # Su rgery i f there is intes tinal p erforation

Enteric fever treatment Chose from the following treatment options for enteric fever

Figure 4.17: Output for Treatment of Enteric fever of the MDES Figure 15:: Output for Treatment of Enteric fever of the ENTERASE

Do you want to end the program

Logout confirmation

? Do you want to end the program

Yes No

Figure 4.8: Log out pop out of the MDES Figure 16: Log Out pop out of the MALARES

especially in remote rural areas of Nigeria and Af-rica, where it is difficult to refer out patients. So Medical diagnosis will have greater part of the ad-vantages of expert system, knowing that only a few medical experts (or specialists) exists in the medical field. The knowledge of such specialists can be rep-licated and made use of in times when they are ex-tremely necessity. The development of the ENTERASE for en-teric fever is a contribution towards reducing deaths due to enteric fever, causing preventable death. This is because its designed to enhance the diagnosis of enteric fever, one of the diseases constituting the leading causes of death although mostly preventable in the developing countries like Nigeria. Therefore are not going to be missed data or misdiagnosis and so will not cause any preventable death. That is to say, by allowing for more efficient diagnosis of en-teric fever, the ENTERASE would aid in reducing the workload of scarce medical practitioners.

References [1] Nkanginieme K.EO. (1997). Clinical diagnosis as a dynamic cognitive process: Application of Bloom’s tax-onomy for educational objectives in the from cognitive domain Med Educ Online [serial online] 1997;2:1. Avail-able: URL http://www.utmb.edu/meo/. [2] Nkuma-Udah K. I. et al (2020), Design of Expert Sys-tem for Diagnosis of Malaria Using VB 6.0 , Afr J Med Phy, Biomed Eng & Sc, 2020, 7, 25 — 37 [3] Frampton S., Guastello S., Brady C., Hale M., Horowitz S. Smith SB. and Stone S. (2008). Patient-Centred Care Improvement Guide. Planetree, Inc, Derby, USA [4] Erin P. Balogh, Bryan T. Miller, and John R. Ball, (2015). Improving Diagnosis in Health Care. Academic Press, available at http://nap.edu/21794 [5] Carter J., Müller-Stöver I., Östensen H. and Heuck C.C., (2005), Good Clinical Diagnostic Practice, WHO Regional Publications, Eastern Mediterranean Series, Cairo, Egypt. ISBN : 978-92-9021-393-2 [6] Igor W., (2001), Design and Implementation of a Graphical User Interface for Computer Aided Logical Design of Kheops Knowledge Based System. MSc The-sis. [7] Nkuma-Udah K.I., Chukwudebe G.A. and Ekwon-wune E. (2019). CLIPS: An Expert System Tool for


Medical Applications, Afr J Med Phy, Biomed Eng & Sc. 6, 61 - 70, retrieved on 04/11/2019 from http://www.ajmbs.abcafrican.com/ nkuma_udah_fullpaper_6_2.pdf [8] Ekenna O. (1992). Typhoid fever: Problems of accu-rate laboratory diagnosis and antimicrobial therapy. Niger Med J, 23: 93-9 [9] Ngwu BA. and Agbo JA. (2003). Typhoid fever: Clinical diagnosis versus laboratory confirmation. Niger J Med, 12: 187-92. [10] Edino ST, Mohammed AZ, Uba AF, Sheshe AA, Anumah M, and Ochicha O. (2004). Typhoid enteric per-foration in north western Nigeria. Niger J Med; 13: 345-9. [11] Sherwal BL, Dhamija RK, Randhawa VS, Jais M, Kaintura A, and Kumar MA., (2004). A comparative study of typhidot and widal test in patients of typhoid fever. J Indian Acad Clin Med, 5: 244-6 [12] Bhan MK, Bahl R. and Bhatnagar S. (2005). Ty-phoid and Paratyphoid fever. Lancet, 366: 749-62 [13] Centres for Disease Control and Prevention CDC, (2014). Typhoid Fever, CDC, USA. Retrieved on 25/12/2017 from https://www.cdc.gov/typhoid-fever/sources.html [14] World Health Organization (CTD/FIL(DEN)/IC/96.1). Accessed on 30th July, 2018 @ http://www.who.int/csr/resources/publications/dengue/whocdsdenic 20001.pdf [15] Enabulele S. and Awunor1 S.N. (2016), Typhoid fever in a Tertiary Hospital in Nigeria: Another look at the Widal agglutination test as a preferred option for di-agnosis, Niger Med J ., May-Jun; 57(3): 145–149. Re-trieved 15th Sept, 2017 @ https://www.ncbi.nlm. nih.gov/pmc/articles/PMC4924394/ [16] Anna E. N. (2014). 3 Infectious Diseases Related To Travel. CDC Health Information for International Travel 2014 : The Yellow Book [17] Addis M. and Sisay D. (2015). A Review on Major Food Borne Bacterial Illnesses. J Trop Dis 3:176. doi:10.4176/2329-891X.1000176 [18] Schulte, M., & Hensel, M. (2016). Models of intesti-nal infection by Salmonella enterica: introduction of a new neonate mouse model. F1000Research, 5, F1000 Faculty Rev–1498. http://doi.org/10.12688/f1000research.8468.1 [19] John A.C. (2017). Epidemiology and global burden of disease of typhoid fever, WHO SAGE: typhoid vac-cines session, retrieved on 26/12/2017, from http://www.who.int/ immunization/sage/meetings/2017/october/2_Crump_Typhoid_SAGE_16Oct2017.pdf

[20] Nkuma-Udah K.I., Chukwudebe G.A. and Ekwon-wune E. (2018). Medical Diagnosis Expert System for Malaria and Related Diseases for Developing Countries, E-Health Telecommunication Systems and Networks 7, 43-56. https://doi.org/10.4236/etsn.2018.72002 [21] Baba A. and Ahmed M.Y. (2013) – Evidence of Ar-bovirus Co-infection in Suspected Febrile Malaria and Typhoid Patients in Nigeria. J Infect Dev Ctries, 7(1):051-059. [22] Ayukekbong, J.A., (2014). Dengue Virus in Nigeria: Current Status and Future Perspective. British Journal of Virology, 1(3): 106-111. [23] Chaudhary N. and Newton F. (2016), Comparison of Hematological Parameters in Various Acute Febrile Ill-nesses. National Journal of Laboratory Medicine. 2016 Jul, Vol-5(3).


[24] Wikipaedia, (2018). Medical Diagnosis. Accessed on the 1st August, 2018 @ https://en.wikipedia.org/wiki/Medical_diagnosis

[25] Carter J., Müller-Stöver I., Östensen H. and Heuck C.C., (2005), Good Clinical Diagnostic Practice, WHO Regional Publications, Eastern Mediterranean Series, Cairo, Egypt. ISBN : 978-92-9021-393-2

[26] Anigbogu S. O. (2006). Artificial Intelligence-Based Medical Diagnostic Expert System For Malaria And The Related Ailments. International Journal of Computer Sci-ence & Applications - IJCSA.

[27] Tunmibi S., Adeniji O., Aregbesola A., and Dasylva A. (2013) “A Rule Based Expert System for Diagnosis of Fever”, International Journal of Advanced Research, 1(7), pp. 343-348.

clinical diagnosis of enteric fever by computer expert systems

Documents