Analysis Of K Means Clustering And Classifiers In Diagnosing Abnormality Of The Ultrasonic Liver Images

This Paper Investigates The Diagnosis Of Liver Abnormalities From UltrasonicImages Using K Means Clustering Algorithm And Five Classifiers. Nowadays, Liver CancerIs One Of The Most Serious Health Problems. Medical Imaging Is Powerful Tool ForDiagnosing The Abnormalities In The Earlier Stage. The Features Are Extracted Using KMeans Clustering And Principal Component Analysis (PCA),Expectation Maximization(EM),EM PCA, Kernel PCA,Gaussian Mixture Model (GMM) Classifier Are Used To DetectThe Liver Image As Normal Or Abnormal. The Parameters Such As Sensitivity, Specificity,Accuracy, Precision, Error Rate, Mathew Correlation Coefficient (MCC), And ClassifierSuccess Index (CSI) Are Analyzed And Compared. When Compared With All TheClassifier The Logistic Regression Attained A Higher Accuracy Of 80.95%.