Literature survey Lung CancerIn this paper, the risk of getting lung cancer willbe acquired and patients will be given directions to eliminate the risk.
Aftercomputing the risk value for lung cancer, status of the patient’s tendency and resistanceto stress is utilized in determining the outcomes of stress to disease. Withrespect to resolve the problem, the neuro-fuzzy logic model has been presentedby author. The system will configure a pre-diagnosis for the people who possiblycan have risk of causing cancer due to living standards and working conditions.Therefore, this will authorize these people to take precautions to reduce of the risk of cancer. A new t-norm operator hasbeen utilized in the problem. Besides the contribution of neuro-fuzzy logicmodel in the field of artificial intelligence and topics of health will also bedetermined by author. A new t-norm operator, which is called MEP (ModifiedEinstein Operator), has been granted and used in the problem. The neuro-fuzzy model for lung cancerhas been advanced.
Age, geneticstatus, details of smoke, nutrition habit, vocation, socioeconomicstatus, living environment and race as factors have been designated in the modelof lung cancer. In harmony with these factors, the membership functions eachand every factor have been resolved. Genfis1, genfis3 and genfis2 functions ofMatlab had used for consistent and stabledevelopment. The system has created by these functions. The most lasting resultwas provided by genfis3 function. This five-input system has been acquired withthe two membership functions for every input and 180 rules.
By this way, theoptimal membership functions and rules of the system had been developed by thesupport of fuzzy FCM–C means explanationgenfis3 functions by using clustering algorithm. The risk result for model havebeen obtained by applying Neuro-Fuzzy method. ANFIS, Einstein product, and theproposed operator have been implemented respectively. As a result of outcomefrom these factors, the risk status of a person to that type of cancer has beendetected within the model.
Breast CancerIn this paper, author exhibit an agent-based systemfor dispersed risk assessment of breast cancer development excercising fuzzyand probabilistic computing. The suggested fuzzy multi agent system coincidesof multiple fuzzy agents that advantage from fuzzy set theory to substantiatetheir soft information (linguistic information). Fuzzy risk assessment isevaluated by two linguistic variables of low and high . Through fuzzyapproximations, the multi agent system valuates the fuzzy probabilities ofbreast cancer development according to various risk factors. As such ranking oflow risk and high risk fuzzy probabilities, the multi agent system (MAS)decides whether the risk of breast cancer advancement is high or low. Thisinformation is then supply an insurance premium adjuster in order toaccommodate preventive decision making in addition to make appropriatemodification of insurance risk and premium.
This final step of insuranceinterpretation also provides a differential measure to demonstrate theconvenience of the approach. Furthermore, actual data collected from twohospitals in Mashhad whilst 1 year. The results are then contrasted with afuzzy distributed approach. Theintroduced fuzzy-probabilistic multi-agent system (MAS) for BC riskcharacterization consists of several risk factor agents, each of the risk factoragent represents one of the BC risk factors. There is also one interface agentwhich plays the role of an interface between user agent and risk factor agents.In this system each risk factor agent is provided with soft and statistical data.
Therefore any risk factor agent that interface to the ith risk factor isenhanced with soft data and statisticaldata nearly with there own risk factor. In fuzzy-probabilistic MAS method, LRand HR women who must pay severally 3.9395e+005 and 0.98285e+007 for their premiumare more desirous to purchase premium ever since suggested premium isconsiderably lower in resemblance with fuzzy MAS in which corresponding premium rates are 1.1069e+007 and 0.
95186e+006 appropriately. Consequentlybecause of proceeding inquiry andcomparisons, the insurance company can discover the fuzzy-probabilistic MAS methodis furthermore accurate for premium assignment and breast cancer premium assignmentin a certain inhabitant, where the method guarantees the bilateral benefits ofthe insured women along with insurance company. Additionally the insurancecompany can ascertain its treatment, screening and prevention services based on the fixed LR and HR insurancepremiums that are intended by the fuzzy-probabilistic MAS. It should bedesignated that the suggested fuzzy-probabilistic multi-agent system assumesFII = 5 in its estimations, where decreasing it to FII = 3 leads to theincrease of LR and HR premiums and subseqently decreases the various insurancepurchasers and increases the prevalent revenue of insurance company. Incontrast, increasing FII to 7 leads to the decrease of premium rates andgeneral revenue of insurance company, whilst attracting further insurancepurchasers. This is although the success average of the author introducedmethod in all these three choices of FII is identical. MASfor those of the fuzzy MAS. The results denote the superiority of the authorsuggested method in the case of breast cancer risk estimation and premiumassignment.
The system can also be enhanced as a social or private healthsystem for offering preventive advices and assessing BC risk notices. Regardless of the financial aspects,the author advised system can be exercised as a private health system hardlyany woman or as a public health system for any professional. The system can beadvanced by offering dissimilar preventive notices for any user almostconvenient life style, screening and health habits ,preventive methods.They provide integrated risk estimate model recommendsthe patient to be recommended to Annual clinical breast analysis and ScreeningMammogram. The authoroffered cognitive mapping situated decision system support not only comprisesthe outcomes of mammographic screening and assessment of demographic risk studybut might also be utilized to help clinical oncologists to discourse various “what-if”questions to investigate the effect of alters in various risk factors on thefinal risk grading. studyaimed at presenting a first attempt to develop an incorporated decision supporttechnique for the estimation of BC risk grade contemplating both demographic andmammographic screening characteristics. Integratedrisk assessment of BC model attaches Fuzzy Cognitive Maps (as the core decisionmaking methodology) in a second level structure: the level-1 FCM models thedemographic risk profile and is trained with the nonlinear Hebbian learningalgorithm to help on predicting the BC risk grade based only on the 14 personalBC risk factors identified by domain experts, and the level-2 FCM models thefeatures of the screening mammogram concerning normal, benign and malignantcases.
The data driven Hebbian learning algorithm is used to train the L2-FCMfocused on the prediction of a new BC risk grade based solely on thesemammographic image features. An overall risk grade is calculated by combiningthe out comes of these two FCMs. Oral CancerOral cancer is the leading cancer type in men andthe third most common cancer in women 2. In India, oral cancer is usuallydetected at advanced stages and the five year survival rate for advanced oral canceris very low 3, the two oral precancers were chosen due to their definitecause–effect relation with the tobacco and related product, areca nut andrelated materials.
this study, intends to apply a fuzzy rule – base for betterprediction of malignancy or pre-malignancy susceptibility viz. OLK, OSF andOSCC other oral disease from as well as mathematical validation ofconsideration of physician’s assumptions and conclusions of previousepidemiological studies in disease prediction chances assigning fuzzy rule base. Fuzzy rule-base approach has beenutilized for value addition to the findings from conventional statisticalapproach in defining particular association between significantclinicoepidemiological parameters and their plausible impact on disease output ina particular dataset. Low literacy rates in conjunction with debilitatingaddictive habits were found to be important underlying reasons for oralprecancers and OSCC occurrence in the studied population. Further, oral healthand habits’ trend analysis through fuzzy If-Then rule demonstrated gender baseddifferences in the awareness outlook in different age groups. Chances ofdisease susceptibility in certain condition can also be predicted by theproposed methodology. The novelty of the proposed approach relies uponconsideration of uncertainty of conditions associated with disease occurrenceand incorporation of physician’s intuition in real-life situations, in contrastto conventional statistical method which predicts disease chances in rigidquantitative values. Hence,the proposed oral pre-cancer, cancer and other oral diseases susceptibilityassessment methodology with embedded fuzzy analytical dimensions depicted theassociation of multiple clinico-epidemiological parameters (viz.
oral healthand literacy as well as addictive oral habits) in simple linguistic terms whichnot only were useful for clinical users but also carried translational values.EsophagealcancerEsophageal cancer is one of the most common cancers world-wideand also the most common cause of cancer death. In this paper, we present anadaptive fuzzy reasoning algorithm for rule-based systems using fuzzy Petrinets (FPNs), where the fuzzy production rules are represented by FPN. Wedeveloped an adaptive fuzzy Petri net (AFPN) reasoning algorithm as a prognosticsystem to predict the outcome for esophageal cancer based on the serumconcentrations of C-reactive protein and albumin as a set of input variables.The system can perform fuzzy reasoning automatically to evaluate the degree oftruth of the proposition representing the risk degree value with a weight valueto be optimally tuned based on the observed data. In addition, theimplementation process for esophageal cancer prediction is fuzzily deducted bythe AFPN algorithm. Performance of the composite model is evaluated through aset of experiments.
Simulations and experimental results demonstrate theeffectiveness and performance of the author proposed algorithms. A comparisonof the predictive performance of AFPN models with other methods and theanalysis of the curve showed the same results with an intuitive behavior ofAFPN models. The author proposedalgorithm of AFPN implemented on the developed experimental system can estimateesophageal cancer based on the serum concentrations C-reactive protein (CRP)and albumin as input variables accurately. The model of AFPN reasoningalgorithm exposes that the model was a powerful tool to combine medicalexperts’ knowledge into a prognostic model based on input data.Existing systemThe detection of risk for people who possibly canget cancer by applying the author usesNeuro-Fuzzy Logic Model and ANFIS. Where ANFIS is utilized to solve the problem is its abilityof effective inference despite uncertain verbal data as human does, as well asgiving better accurate results even in nonlinear problems with the help of thelearning ability of the model and ANFIS binds input and output variables usingthe learning ability of artificial neural networks and sets fuzzy rules24.
The system converts itself into having the capability of learning by usingtraining set for fuzzy modeling procedure. ANFIS unifies the decision abilityand verbal expression of fuzzy logic with help of ability to learn andadaptation of neural network. The methodput forwarded in this work is concluded by reforming the fuzzy operators inANFIS. The risk result for model has been obtained by using Neuro-Fuzzy method.ANFIS, Einstein product, and the proposed operator have been suggested byauthor.Inorder to do risk inspection and prevention of Breast Cancer (BC) the authorsuggested a approach of fuzzy cognitive maps (FCMs) which is effectively typifyhuman experience and knowledge, introduces concepts to illustrate the essentialelements and the cause-and-effects.
The decision making system is constructedto estimate risk factors of BC according to knowledge for oncologists. Theresolution of breast cancer (BC) risk factors is essential because it persuadethe protective and preventive treatment plans of patients In this paper, Rule-Based FuzzyCognitive Map (RBFCM) approach is utilized, which is a robust tool for modelingof dynamic systems to represent and provide a model for experts’ knowledge. RBFCMcan successfully display the knowledge and experience of the experts.Oralcancer is commonly detected at advanced stages and the five year survival ratefor advanced oral cancer is very low hencethis also plans to apply a fuzzy rule – base for superior prediction of pre-malignancysusceptibility or malignancy viz. OLK, OSF and OSCC other oral disease alongwith mathematical validation of consideration of physician’s opinions andconclusions of previous epidemiological studies in disease prediction chancesassigning fuzzy rule base.
It would further help the health care giver toforetell chances of disease causes public health prevention efforts. Disease sensitivity was appraised usingformed If-Then rules in specific conditions. The rules were also inferred forpredicting oral disorder other than oral cancers (OTH) and precancers. Susceptible topology and collaborative structure of theMASs makes them applicable in large area applications.
This paper initiate acooperative MAS where agents can share knowledge and dispense subtasks. By using this strategy they gainfrom self-organization and decentralization helps to solve diffused problems more effectively. The system, agentsare posses with suitable computational capabilities and knowledge. The approachused here is fuzzy-probabilisticmulti-agent system (MAS) for BC risk estimation includes several agents riskfactor this agents includes one of the BC risk factors. The author advisedfuzzy multi agent system coincides of multiple fuzzy agents that utility fromfuzzy set theory to establish their soft information (linguistic information).Fuzzy risk assessment is measured using two linguistic variables of high andlow. Through fuzzy calculations, the multi agent system accounts the fuzzyprobabilities of breast cancer development based on several factors of risk.
By ranking of fuzzy probabilities based onhigh risk and low risk the multi agent system (MAS) considers whether the breastcancer risk development is low or high. Thusthe assessed risk is applied to calculate the insurance company suggests preventive facilities when the user with acertain risk.Author introduce an approach for fuzzy probabilities estimationwhich is particularly helpful combine imprecise probabilities and model derivedfrom statistical data and linguistic data By applying fuzzy probabilities, helpsto utilizing parallel structure of MASs at the same time. Hence can save timeand decrease proportional complexities Thenany risk factor utilizes statistical data and soft data where linguistic dataand linguistic facts aboutrisk factors The statisticaldatabase obtained from patients for screening financial services and Preventivetreatment services may partially or completely cover magnetic resonance imaging(MRI), chemoprevention expenses prophylactic ophorectomy ,mammography,mastectomy and other treatment services which should have been considered inthe maximum expenses of BC treatment and prevention Throughand shows how fuzzy logic can be advantaged for modeling soft data.
Result sharing and cooperation with aflexibility and higher reliability.Theauthor employes a technique for modelingof esophageal cancer problem on thebasis of AFPN. The risk factor obtained is real and can explain the risk factors of esophageal cancer as a setof factor where alcohol consumption, exposure of esophageal tissue to acid,possibly hot liquids, unhealthy diet tobacco smoke, on the otherhand technological advances of enhanced diagnosis and therapeutics the forecastfor esophageal cancer remains unsatisfactory Itpresents an adaptive fuzzy Petri net model to resolve the esophageal cancer. A few definitions of FPNthat are needed to understand the modeling capability of AFPN. The AFPNs approach could be a very goodsubstitute for other biologicalprocesses.For flexible and represented fuzzy rules for modeling of all other albuminand CRP and the output of esophagealcancer into illustrating fuzzy sets. the model accomplished in this paper hasone output and twoinputs.
By this way, forany value of CRP, albumin, membership degree is l,and the risk degree based on this information be acquired. Therfore 9 fuzzyrules needed for the reasoning process as a prognostic model of esophagealcancer. However, sometimes it is required to depend on experts to conclude theparameters of the adjustable model Membershipfunctions for output variable ”risk degree” and input variables ”CRP and albumin’. Thus AFPNmodel is able to handle The first production rule derives the possibility thatalteration risk degree level is very low; the second production rule derivesthe possibility that the change risk degree level is low, the third productionrule derives the possibility that the change risk degree level is medium, thefourth production rule derives the possibility that the change risk degreelevel is high, the fifth production rule derives the possibility that thechange risk degree level is very high.
The model of AFPN conviction algorithmexposes that model is a powerful tool to merge prognostic model Into medicalexperts’ knowledge on the basis of input data.