Medical Expert Systems For Diagnosis Of Various Diseases
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014Medical Expert Systems for Diagnosis of VariousDiseasesJimmy SinglaDinesh Grover, PhDAbhinav BhandariResearch ScholarIKG PTUEx-DirectorDept. of CSE & ITLLRIET, MogaAsst. Prof.Dept. of CE, UCOEPunjabi University, PatialaABSTRACT2. EXPERT SYSTEMDiseases should be treated well and on time. If they are nottreated on time, they can lead to many health problems andthese problems may become the cause of death. Theseproblems are becoming worse due to the scarcity ofspecialists, practitioners and health facilities. In an effort toaddress such problems, studies made attempts to design anddevelop expert systems which can provide advice forphysicians and patients to facilitate the diagnosis andrecommend treatment of patients. This review paper presentsa comprehensive study of medical expert systems fordiagnosis of various diseases. It provides a brief overview ofmedical diagnostic expert systems and presents an analysis ofalready existing studies.Expert System is one of the most common applications ofartificial intelligence. It is a computer program that simulatesthe decision and actions of a person or an association that hasspecialist facts and experience in aparticular field.Normally, such a system contains a knowledge basecontaining accumulated experience and a set of rules forapplying the knowledge base to each particular situation. Themajor features of expert system are user interface, datarepresentation, inference, explanations etc. Advantages ofexpert system are increased reliability, reduced errors, reducedcost, multiple expertise, intelligent database, reduced dangeretc. Disadvantages of expert system are absence of commonsense and no change with changing environment.[9].General TermsA Fuzzy Expert System is a group of membership functionsand rules. These functions and rules are used to reason aboutdata. Fuzzy expert systems are oriented toward numericalprocessing. It takes numbers as input, and then translates theinput numbers into linguistic terms like Small, Medium andlarge. Then the task of Rules is to map the input linguisticterms onto similar linguistic terms describing the output.Finally, the translation of output linguistic terms into anoutput number is done [9].Artificial intelligence, Expert system, Medical knowledge.KeywordsDisease, Symptom, Rule, Patient.1. INTRODUCTIONArtificial Intelligence is defined as brainpower exhibited byan artificial unit. It is a division of computer science dealingwith sharp behavior, knowledge. Research in artificialintelligence is anxious with producing machines tocomputerize jobs requiring sharp actions. Examples includecapability to answer diagnostic and user question, speech andfacial recognition [4]. Artificial intelligence is separated intotwo categories. These two categories are conventionalartificial intelligence and computational intelligence.Conventional artificial intelligence includes machine learningand stastical analysis. Computational intelligence includesneural networks and fuzzy systems. The other applications ofartificial intelligence are automation, computer vision,artificial creativity, expert system and knowledgemanagement [6].The rest of the paper is organized as follows. Section 2provides a brief introduction to Expert System, Fuzzy expertsystem, artificial neural network and neuro-fuzzy technique.Section 3 provides introduction to medical expert system andmedical knowledge. Section 4 describes the structure ofmedical expert system. Section 5 describes the working ofmedical expert system. Section 6 presents the related workand a comparative analysis of existing studies. Finally, thediscussion is concluded in Section 7.Crisp Input ValuesFuzzificationFuzzy RuleBaseFuzzy Inference EngineDefuzzificationCrisp Output ValuesFig. 1: Fuzzy Expert System36
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014Artificial Neural Network helps doctors to understandcomplex clinical data across a large number of medicalapplications. In medical application, the task is on the basis ofthe measured features to allocate the patient to one of a smallset of classes. An artificial neural network is a computationalmodel that tries to report for the parallel nature of theindividual brain. An artificial neural network is anarrangement of extremely interconnecting processingelements operating in parallel. These elements are stirred bybiological nervous systems. As in environment, theconnections between basics largely find out the networkfunction. A subgroup of processing part is called a layer in thenetwork. The primary layer is the input layer and the finallayer is the output layer. Between the input and output layer,there may be extra layers, called hidden layers [10].Input Valuesrules is prepared in the knowledge base. You may understandit with the help of example. Tuberculosis is a lung diseasewhose symptoms are persistent cough, constant fatigue,weight loss, loss of appetite, fever, coughing up blood, nightsweats. So it will be stored in knowledge base in the form of arule which is as follow:Disease (Patient, tuberculosis):Symptom (Patient, persistent cough),Symptom (Patient, constant fatigue),Symptom (Patient, weight loss),Symptom (Patient, loss of appetite),Symptom (Patient, fever),Symptom (Patient, coughing up blood),Symptom (Patient, night sweats).InputLayerFig.3: Production Rule in Knowledge BaseSimilarly in this way you can store maximum possible rules inthe knowledge base.HiddenLayer2) The Fact Base contains facts which are applied to match inopposition to the antecedent part of rules stored in theknowledge base. The fact base is analogue to the instanthuman memory.3) The foremost job of Inference Engine is to bring out thereasoning by connecting the rules with facts and deducingnew facts.OutputLayer4) The User Interface is used to correspond among user andexpert system.Output ValuesFig. 2: A typical Neural Network [10]5) The Explanation Module permits the user to inquire theexpert system how a finicky conclusion is reached and why aspecific fact is desired.A Neuro-fuzzy System is a fuzzy system that uses a learningalgorithm inspired by neural network theory to determine itsparameters by processing data samples [14].6) The Developer Interface is used to alter the knowledge base3. MEDICAL EXPERT SYSTEMIn the fig. 4 the simulation of medical expert system ispresented. In the figure, S1 D1 denotes the first symptom offirst disease. In general Si Dj denotes the “i” symptom of “j”disease. If the program has a positive answer to the symptom,it goes on with the symptoms from that disease. If a symptomfrom that disease is negative, it jumps to the first symptomfrom the next disease.A huge figure of expert systems is medical. The chief aim ofany medical expert system is identification and cure ofdiseases. A medical expert system is built up of programs andmedical knowledge base. The information obtained frommedical expert system is similar to the information given byproficient in that particular area [6].Medical Knowledge of specialized doctor is vital for thegrowth of medical expert system. This knowledge iscomposed in two phases. In the first phase, the medicalconditions of diseases are recorded during the formation ofpersonal meeting with doctors and patients. In the secondphase, a deposit of rules is formed where each rule contains IFpart that has the symptoms and THEN part that has thedisease that should be realized [12].4. THE STRUCTURE OF MEDICALEXPERT SYSTEMSA medical expert system has the following components1) The Knowledge Base encloses information with referenceto diseases which are characterized as a set of if-thenproduction rules. The knowledge base is analogue to the longstanding human memory. The whole sorting of production5. WORKING OF MEDICAL EXPERTSYSTEMS6. RELATED WORKSThe Medical diagnostic Systems have undergone manychanges and are using new techniques to generate betterresults. This section briefly summarizes some of the fuzzylogic, rule based and artificial neural network based medicaldiagnostic systems and presents a comparative analysis ofvarious existing studies in table 5.Mohammed Abbas Kadhim, M. Afshar Alam, Harleen Kaur[9] build up a fuzzy expert system for judgment of back paindiseases based on the experimental examination symptomsusing fuzzy rules. The user has to key in parameters such asbody mass index, age, gender of patient and experimentalexamination symptoms for this fuzzy expert system. On thebasis of these parameters, this fuzzy expert system makessuitable judgment of back pain diseases and gives somemedical suggestion to the patient. The accuracy achieved fromthis fuzzy expert system was 90%.37
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014NONOS1 D2S1 D1NOYESNONOYESThe other D1symptomsYESS1 D3The other D2symptomsYESYESResult is D1NOThe other D3symptomsResult is D2YESResult is D3S1 DiNOS1 DnYESYESNOThe other DisymptomsThe other DnsymptomsYESYESResult is DnResult is DiFig. 4: Working of Medical Expert System [4]38
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014The classification of input variables in this expert system isdone as follow:Table 1: Classification of Input Variables [9]InputVariableRangeAgeBMISymptomFuzzy Sets 25Young22-40Middle38 Old 18Low19-25Medium24 Hgh 20Low18-60Medium58-80High78 Very HighQeethara Kadhim AI-Shayea [10] evaluate artificial neuralnetwork in diagnosis of diseases. Two cases were considered.The first one was acute nephritis disease; data was the diseasesymptoms. The second was the heart disease; data was oncardiac Single Proton Emission Computed Tomographyimages. Each patient classified into two categories: infectedand non-infected. Categorization is an important tool inmedical diagnosis decision support. Feed-forward backpropagation neural network is applied as a classifier todistinguish between infected or non-infected person in bothcases. In the diagnosis of acute nephritis disease; the percentcorrectly classified in the model sample by the feed-forwardback propagation network is 99 percent while in the finding ofheart disease; the percent correctly classified in the modelsample by the feed-forward back propagation network is 95percent. The analysis variables of data set used in this studyare shown in table 2Table 2: Diagnosis Variables of data set [10]Eugene Roventa, George Rosu [4] enlarged an expert systemwhich is applied to detect major kidney diseases. Thediagnosis is made using the readings obtained from clinicalexam and the paraclinical exam. This system helps themedical expert in making the suitable analysis of a patient. Alot of common symptoms often occur in kidney diseases andmany of them are alike, and that makes it difficult even for akidney doctor to place an exact diagnosis. This expert systemeliminates this trouble. This expert system has a very wellbuilt knowledge base. It has knowledge of twenty sevendiseases from nine dissimilar categories.K.Abdelhamied, S.Hafez, W.Abdalla, H.Hiekal, A.Adel [8]developed an expert system to aid health employees workingin crowded outpatient clinics in Egypt make quick and perfectjudgment. In this system, the medical knowledge of more than300 major and minor common diseases is encoded in the formof production rules. These rules are invoked in collectionsaccording to one or more presenting symptoms and diagnostichypotheses are devised for the diseases most expected toreason these symptoms. Once a disease is identified,recommendation is given for cure. The system will beassessed in 10 outpatient clinics.Solomon Gebremariam [13] built up a prototype self-learningknowledge-based system that can offer advice for physiciansand patients to assist the diagnosis and cure of diabeticpatients. Knowledge is obtained using both structured andunstructured interviews from domain experts which arechosen using purposive sampling technique from Black LionHospital Diabetes Center. It brings into play backwardchaining which initiates with possible solutions and tries toassemble information that verifies the result. The general totalperformance of the prototype system is 84.2%.Theassessments taken in this expert system are shown as follow:Table 3: Decision taken in Expert System [13]SymptomsNon Related SymptomsResultLess Likelyhave DiabetestoRelatedSymptomsLab Result usingFPG 100 – 125mg/dLRelatedSymptomsLab Result usingFPG 99 mg/dLDiabetes FreeRelatedSymptomsLab Result usingFPG 126 mg/dLType 1 / Type 2 /other DiabetestypePrediabetesDiagnosis Variable NameTemperature of Patient {35 C – 42 C}Occurrence of Nausea {Yes or No}Lumbar Pain {Yes or No}Urine Pushing {Yes or No}Micturition Pains {Yes or No}Burning of Urethra, itch, swelling of urethra outlet{Yes or No}Dr. Sandeep Pachpande, Ramesh Mahadik [2] planned andput into operation a medical expert system for the diagnosis ofsome Pulmonary Disorders. One of the endings accomplishedthroughout the route of this research was that the process ofknowledge acquisition is a continuous process and for thisreason an expert system cannot be made in a single passfashion – an incremental approach is certain.S Ali, P Chia, K Ong [11] present an automated deliverysystem for clinical guidelines (DSCG) that supports cliniciansin diagnosing and treating patients bearing from chest pain inthe emergency department. Strategies are adaptively selectedfrom a knowledge base server that has a library of clinicallydefined, graphical guidelines. The system acquires patientdata, such as illness and assessment results, and matches these39
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014data with eligibility criteria. It recommends most favorabletreatment plans and analysis based on the most feasiblediagnosis. Clinicians may either use the recommendations as asuggestion or trigger a selection to check the patient’scircumstance during the cure using an intelligent agent.Ede Kekes, Istvar Laczay, Janos Barcsak, Peter Kcch, ZoltanAntaloczy [3] developed an expert system for the valuation ofthe analysis and therapy ischemic heart disease in a decisionmodel. It works in 2 different manners:1. As a system facilitating the effective work or screening thepopulation for IHD.2. As an instructing model in the university and postgraduatemedical education schools. It is executed in turbo prolog.John G. Holmen, Anthony H. Walff [7] proposed an expertsystem that offers advice about oliguria arising on theIntensive Care Unit. Justifications are raised by discoveringprobable paths in a causal network of physiological states.The expert system shell expanded for this purpose is writtenin Prolog and runs on the IBM PC. Investigative deeds arescheduled using an schema and may be suspended andcontinued as required. Although the shell is wide-ranging, itmaintains the definition of knowledge representationlanguages tailored to a fussy application. This competence isapplied to build a frame notation for the representation ofphysiological states.Freasier, R.E, Cios, K.J, Goodenday, L.S [5] constructed amedical expert system using the multiplication model ofsupport logic programming so as to establish the predominantstenosis in one of the three main coronary arteries. Thefeatures taken in the determination of the analysis were datagot from preprocessed scintigraphic myocardial perfusionimages of the left ventricle in use in three views: anterior, leftlateral and left lateral frontal oblique. Stress thallium-201planar scintigrams from long-sufferings who had coronaryarterial stenosis confirmed by coronary arteriography includedthe data set used for this reading. A prolog-based system wasput up to discover the knowledge base so as to find out thesite of the predominant stenosis. With the current set ofproduction rules, the system properly recognized the site ofcoronary artery stenosis in over 90% of the long-sufferingstaken into account.Jimmy Singla [6] developed an expert system to identify themost important lung diseases between the patients. Thejudgment is made using the symptoms that can be felt by thepatient. This medical expert system aids the doctor or expertin building the proper diagnosis of the patient. The lungdiseases have many regular symptoms and some of them arevery much alike. This creates much complicatedness for thephysician to reach at a right conclusion. This expert systemcan take away this complicatedness and it is havingacquaintance of thirty-two lung diseases. Its correctness is70%.Samy S. Abu Naser, Abu Zaiter A. Ola [12] developed anexpert system that urges the patient with conditions forsuitable analysis of some of the eye diseases. The eye hasalways been viewed as a tunnel to the inner workings of thebody. The disease states frequently generate symptoms fromthe eye. CLIPS language is used as a tool for drawing expertsystem. A preliminary evaluation of the expert system waspassed out and a optimistic response was acknowledged fromthe users.Ahmad A, Al-Hajji [1] presented a Rule-Based Expert Systemfor Neurological Disorders. This system diagnoses and treatsmore than 10 types of neurological diseases. It helps thepatients to acquire the required recommendation regarding theunusual disorders attack to them due to their nervous systemdisorders. The expert rules were built up on the symptoms ofeach type of neurological disease, and they were offered usingdecision tree and deduced using backward-chainingtechnique. The knowledge base contains information,gathered from volumes and practitioners about neurology andits disorders.Table 4: Diseases diagnosed in Expert System [1]S No.Name of Disease1Meningitis2Cerebral palsy3Migraine4Cluster headache5Stroke6Epilepsy7Multiple sclerosis8Parkinson9Alzheimer10Huntington diseaseTable 5: Comparison of Existing Studies on Medical Expert SystemsReferenceMohammedAbbasKadhim, M. AfsharAlam, Harleen Kaur [9]DiseaseDiagnosedBack Pain DiseaseTechnique UsedFuzzy ExpertSystemInputRemarksBody mass index, age,gender and clinicalobservation symptomsThis system is tested usingclinical data that corresponds to20 patients with different backpain diseases and 90% accuracy isachieved.40
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014ReferenceDiseaseDiagnosedTechnique UsedAcute NephritisDiseaseSolomon Gebremariam[13]SandeepPachpande,Ramesh Mahadik [2]Disease symptoms99% accuracy is achieved indiagnosis of Acute NephritisDisease.Data is on Cardiacsingle proton emissioncomputed tomographyimages95% accuracy is achieved indiagnosis of heart disease usingfeed forward back propagationnetwork.Disease symptomsThis system contains knowledgeof 27 kidney diseases but noexperimental results are found.Disease symptomsThe system contains knowledgeof 300 major and minor diseases.It is being evaluated in 10outpatientclinicsbutnoexperimental results are given.Disease Symptoms, Labtest results, Age, Familyhistory, obesity, KetoneThis system provides advice tophysicians and patients tofacilitate the diagnosis andtreatmentofdiabetesthePerformance of the system is84.2%.Rule based ExpertSystemDisease SymptomsThis expert system shows thatconstruction of expert system isnot a single pass fashion. It is anincrementalapproach.Noexperimental results of this expertsystem are found.Heart DiseaseK.Abdelhamied,S.Hafez,W.Abdalla,H.Hiekal, A.Adel [8]RemarksArtificial NeuralNetworkQeethara Kadhim AIShayea [10]EugeneRoventa,George Rosu [4]InputKidney DiseaseMajor and MinorDiseasesDiabetes DiseasePulmonarydisordersRule based ExpertSystemRule based ExpertSystemRule based ExpertSystemS.Ali, P. Chia,K Ong [11]Chest PainKnowledge BasedSystemData obtained fromLaboratoryExaminations, Chest XRay images, UltrasoundVideo, Narrative textsdescribing the patient'sconditionJohnG.Holmen,Anthony H. Walff [7]Oliguria occurringon the IntensiveCare UnitKnowledge BasedSystemCentral VenousPressureThis expert system deliversappropriate clinical guidelines andis finalized for pilot trial at theaccidentsandemergencydepartment of the nationaluniversityhospital.Noexperimental results of this expertsystem are found.This system gives advice aboutoliguria occurring on intensivecare unit and no experimentalresults are shown.41
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014DiseaseDiagnosedReferenceTechnique UsedRemarksData obtained frompreprocessedscintigraphicmyocardial perfusionimages of the leftventricle taken in threeviewsThis system determines the site ofthe predominant stenosis. Withthe current set of production rules,the system properly recognizedthe site of coronary artery stenosisin over 90% of the patientspresented.This expert system containsknowledge of 32 lung diseasesand the system has70% accuracy.Freasier, R.E, Cios, K.J,Goodenday, L.S [5]PredominantCoronary ArterialStenosisKnowledgeSystemJimmy Singla [6]Lung DiseasesRule based ExpertSystemDisease symptomsSamy S. Abu Naser,Abu Zaiter A. Ola [12]Eye DiseasesKnowledgeSystemBasedDisease symptomsAhmad A, Al-Hajji [1]NeurologicalDisordersRule based ExpertSystemDisease symptomsObi J.C, Imianvan A.A[16]LeukemiaNeuro–FuzzyExpert SystemDisease symptoms7. CONCLUSIONSThis review paper describes different expert systems inmedical diagnosis and evaluates the contributions made bydifferent researchers. Some researchers have evaluated theirmedical expert systems in hospitals from the experts andretrieved various parameters like accuracy and precision.Using these parameters, they have calculated the performanceof their expert systems. The accuracy and other parameters ofexpert system depend on the knowledge base. The knowledgebase should have relevant knowledge. There should be stresson knowledge acquisition, a stage in which knowledge isgathered. So performance of expert system depends on allthese factors. One can increase the performance of expertsystem by making knowledge base more accurate and verylittle work is done using neuro-fuzzy, ANN and fuzzy logic inmedical diagnosis. So we will go for these in medicaldiagnosis.8. ACKNOWLEDGEMENTSJimmy Singla Author wants to express his sincere thanks toDr. Dinesh Grover, Ex-Director, Lala Lajpat Rai Institute ofEngg. & Tech. Moga , Prof. Abhinav Bhandari, PunjabiUniversity, Patiala , his Parents Er. Rajiv Singla, SubDivisional Engineer & Mrs. Bindu Singla, Er. Nikita, Asst.BasedInputThe proposed system can helpdoctors and patients in providingdecisionsupportsystem,interactive training tool and expertadvice. A number of doctors andpatients tested the system andgave a positive feedback but noparameters are calculated for thisexpert system.This expert system helps thepatients to get the required adviceabout the different disordersattack to them due to theirnervous system disorders. SinceNo parameters are retrieved.This expert system tells thepatient his current condition asregards leukemia. No parametersare retrieved for this expertsystem.Prof., MMU Ambala for guiding him at every step in hiscareer.9. REFERENCES[1] Ahmad A. Al-Hajji, “Rule Based Expert System forDiagnosis and Symptom of Neurological disorders”,proceedings of ICCIT 2012.[2] Dr. Sandeep Pachpande, Ramesh Mahadik, “ExpertSystem for Diagnosis of Pulmonary Disorders”, ASM’sInternational E-Journal of Ongoing Research inManagement and IT, INCON 13-IT-018, pp 01-08.[3] Ede Kekes, Istvar Laczay, Janos Barcsak, Peter Koch,Zoltan Antaloczy, “CORONARIA Expert System forDiagnosis and Therapy of Ischemic Heart Disease”,proceedings of IEEE Engineering in Medicine & BiologySociety tenth Annual International Conference 1988IEEE.[4] Eugena Roventa, George Rosu, “The Diagnosis of SomeKidney Diseases in a PROLOG Expert System”,proceedings of the third international workshop on SoftComputing Applications 2009 IEEE.42
International Journal of Computer Applications (0975 – 8887)Volume 93 – No.7, May 2014[5] Freasier, R. E, Cios, K. J, Goodenday, L.S,“Determination of Predominant Coronary ArterialStenosis by a Knowledge Based System”, proceedings ofIEEE Engineering in Medicine & Biology Society tenthAnnual International Conference 1988 IEEE.[6] Jimmy Singla, “The Diagnosis of Some Lung Diseases ina PROLOG Expert System”, International Journal ofComputer Applications, vol. 78, no. 15, pp. 37-40,September 2013.[7] John G. Holman, Anthony H. Walff, “An Expert Driverfor OLIGURIA occurring on the Intensive Care Unit”,proceedings of IEEE Engineering in Medicine & BiologySociety tenth Annual International Conference 1988IEEE.[8] K. Abdelhamied, S. Hafez, W. Abdalla, H. Hiekal, A.Adel, “ A Rule – Based Expert System for RapidProblem Solving in Crowded Outpatients Clinics inEGYPT”, proceedings of the IEEE Engineering inMedicine & Biology Society tenth Annual InternationalConference 1988 IEEE.[9] Mohammed Abbas Kadhim, M. Afshar Alam, HarleenKaur, “Design and Implementation of Fuzzy expertSystem of Back Pain Diagnosis”, International Journal ofInnovative technology & Creative Engineering, vol. 1,no. 9, pp16-22, September 2011.[10] Qeethara Kadhim AI – Shayea, “Artificial NeuralNetworks in Medical Diagnosis” , IJCSI InternationalJournal of Computer Science Issues, vol. 8, issue 2, pp150-154, march 2011.[11] S Ali, P Chia, K Ong, “Graphical Knowledge – BasedProtocols for Chest Pain Management” proceedings ofthe Computers in Cardiology 1999 IEEE.IJCATM : www.ijcaonline.org[12] Samy S. Abu Naser, Abu Zaiter A. Ola, “An ExpertSystem for Diagnosing Eye Diseases Using CLIPS”,Journal of Theoretical and Applied InformationTechnology, pp. 923-930, 2005-2008 JATIT.[13] Solomon Gebremariam, “A Self Learning KnowledgeBased System for Diagnosis and Treatment of Diabetes”,Master’s thesis, Addis Ababa University, Ethiopia.[14] Arshdeep Kaur, Amrit Kaur “Comparison of fuzzy logicand neuro fuzzy algorithms for air conditioning system”international journal of soft computing and engineering,vol. 2, issue 1, march 2012.[15] S. M. Kamruzzaman, Ahmed Ryadh hasan, Abu Bakarsiddiquee, Md. Ehsanul Hoque mazumder “Medicaldiagnosis using neural network”, proceedings of the thirdinternational conference on electrical and computerengineering December 2004.[16] Obi J.C, Imianvan A.A “Interactive neuro-fuzzy expertsystem for diagnosis of leukemia”, global journal ofcomputer science and technology vol. 11, issue 12,version 1.0, july 2011.10. AUTHOR’S PROFILEJimmy Singla is Assistant Professor at Punjab Institute ofTechnology, Hoshiarpur , a constituent institute of PunjabTechnical University Jalandhar. He has received his B.Techand M.Tech in Computer Science & Engg. From LLRIET,Moga. He is now pursuing Ph.D in Computer Engg. FromPunjab Technical University Kapurthala. His areas of interestare fuzzy logic, expert systems and digital image processing.He has to his credit many papers in national, internationaljournals and conferences and a book on asynchronous transfermode.43
medical expert system. Section 6 presents the related work and a comparative analysis of existing studies. Finally, the discussion is concluded in Section 7. 2. EXPERT SYSTEM Expert System is one of the most common applications of artificial intelligence. It is a computer program that
Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original
10 tips och tricks för att lyckas med ert sap-projekt 20 SAPSANYTT 2/2015 De flesta projektledare känner säkert till Cobb’s paradox. Martin Cobb verkade som CIO för sekretariatet för Treasury Board of Canada 1995 då han ställde frågan
service i Norge och Finland drivs inom ramen för ett enskilt företag (NRK. 1 och Yleisradio), fin ns det i Sverige tre: Ett för tv (Sveriges Television , SVT ), ett för radio (Sveriges Radio , SR ) och ett för utbildnings program (Sveriges Utbildningsradio, UR, vilket till följd av sin begränsade storlek inte återfinns bland de 25 största
Hotell För hotell anges de tre klasserna A/B, C och D. Det betyder att den "normala" standarden C är acceptabel men att motiven för en högre standard är starka. Ljudklass C motsvarar de tidigare normkraven för hotell, ljudklass A/B motsvarar kraven för moderna hotell med hög standard och ljudklass D kan användas vid
LÄS NOGGRANT FÖLJANDE VILLKOR FÖR APPLE DEVELOPER PROGRAM LICENCE . Apple Developer Program License Agreement Syfte Du vill använda Apple-mjukvara (enligt definitionen nedan) för att utveckla en eller flera Applikationer (enligt definitionen nedan) för Apple-märkta produkter. . Applikationer som utvecklas för iOS-produkter, Apple .
3.7 Perceived requirement for expert systems 52 3.8 The role of an expert system 52 3.9 Responsibility of expert systems 55 3.10 Motivation for developing an expert system 56 3.11 Validation 59 3.12 Summary 60 4. EXPERT SYSTEM DEVELOPMENT 4.1 Introduction: A very brief background 61 4.2 Categories of information systems and problems 61 4.3 .
the expert's expertise is "good" and the expert himself (or herself) is ' good"then the Expert System will perform admirably. However, models can never be 100% accurate and no expert is omniscient. Because of this, it is important that users of Expert Systems exercise caution in interpreting the answers produced by these systems.
This chapter is a broad introduction to expert systems. The fundamental princi-ples of expert systems are introduced. The advantages and disadvantages of ex-pert systems are discussed and the appropriate areas of application for expert systems are described. The relationship of expert systems to other me
