In the recent past, the development of computer aided diagnosis systems has been prepared to assisting the physicians for better decision making. This has motivated the research in creating vast amount of image database in the hospitals and health care centers. It has been reported that the brain tumor is one of the major causes leading to higher incidence of death in human. Physicians face challenging task in extracting the features and decision making form. The computerized tomography (CT), which is found to be the most reliable method for early detection of tumors. Due to the high volume of CT images to be used by the physicians, the accuracy of decision making tends to decrease. This has further increased the demand to improve the automatic digital reading for decision making. This paper proposes the tumor detection in brain images. The authors investigate the use of different data mining techniques namely, neural network and association rule mining for anomaly detection and classification. The method proposed makes use of association rule mining technique to classify the CT scan brain images into three categories namely normal, benign and malignant. It combines the low-level features extracted from images and high level knowledge from specialists. The developed algorithm can assist the physicians for efficient classification with multiple keywords per image to improve the accuracy. The results show that the classification accuracy has been obtained as 75% for the classifier using association rule and 70% for neural network classifier, making it a suitable scheme for image mining applications.