Brause (2001) in his interesting study brought to limelight that human beings,however experienced and however enlightened would go wrong in diagnosing disease. The success percentage of his study are as follows: • Best human diagnosis (most experienced Physicians): 79.7%. • Computer with expert data base: 82.2%. • Computer with 600 patient data: 91.1%. This compels to establish the truth that humans cannot ad hoc analyze error-free complex data. Researchers have found that neural network capabilities can help them to improvise this domain. The implementation of human intelligence in scientific equipment has had been the subject of scientific research for a long time and of the medical research in the last decade. This paper carried out to generate and evaluate both statistical and neural network models to predict malignancy of breast tumor,using Wisconsin Diagnosis Breast Cancer Database (WDBC). The objectives in this article are: (i) Compare the diagnostic performance of statistical and neural network models in distinction between malignance and benign patterns,(ii) Reduce the number of benign cases sent for biopsy using the best model as a supportive tool,and (iii) Validate the capability of the best model to recognize new cases. © 2016,International Journal of Pharmacy and Technology. All rights reserved.