Decision Support System For Lung Cancer By using Bayesian Classification

In many application domains, classification ofcomplex measurements is essential in a diagnosticprocess. Correct classification of measurements mayin fact be the most critical part of the diagnosticprocess. The main feature of the proposed system isto provide a sample and integrated tool fordesigning diagnostic application. Lung cancer is thesecond most common malignancy in men and thethird most common cancer in women. Usually lungcancer nodules have a multifocal origin and arather poor prognosis. Therefore, a careful reviewof the symptoms presented and a detailed physicalexam greatly help with the diagnosis occurs. Thispaper proposes a decision support system for lungcancer classification using Bayesian Analysis tohelp the physician or the patient who tests herself athome for lung cancer with the most possible result.Bayesian classification is one of the classificationmethods successfully applied to the cancerdiagnostic problems. The system stores theknowledge of the medical experts and the medicalrecords of the previous cases. Based on theknowledge stored, the system will learn the patternsusing Bayesian Analysis and decide the probabilitybased on the symptoms of the patients. Classifieraccuracy is also estimated to get the better decisionsupport system with the minimum error rate byusing Bayesian Analysis that provides a theoreticaljustification and lower error rate than otherclassifiers, [6]