An Automated System to Predict and Analyze Cancer Supported by Machine Learning Techniques

Abstract

Breaking all bonds of age, sex and condition of a patient, Cancer remains a highlighted disease that requires attention at all times. Screening as a technique to look into shreds of evidence of the existence of cancer tissues proves useful. Though used to a great extent, little thought is given towards the accuracy, efficiency and throughput of systems performing the task. Also data to satisfy these goals is not readily available and to add to it reliability of the data is debatable. There is a highly supported controversy regarding the appropriate age limits for the intended use of screening mammography, leading to ambiguity for screening for a specific organ being meaningful. Therefore, it is required to understand the actual underlying principles of screening and then discuss current controversies.

In this paper, an elaborate discussion is presented on the potential of screening techniques to detect abnormalities in a patient’s body organs. Concentration on abiding by safety and legal aspects is stressed and adoptions of methods that are financially viable are carried out. Machine learning approaches of classification are applied to locally generated data.