Max Pooling Technique to Detect and Classify Medical Image for Ovarian Cancer Diagnosis

Abstract

Machine Learning plays a vital role in the field of Image classification. Researcher’s originate image classification is one of the complex process. As well as researcher’s strongly consider, the possibility of health screening using machine learning is massive and without compromising technical industries and professional healthcare bodies should compose to create an efficient healthcare screening system. Recently, reduction in the production of medicine and drugs, contribution of health screening in identifying diseases by Artificial Intelligence (AI) is essential. Considering various diseases, ovarian cancer is highly rated among women’s. Production of Ovum and hormones is usually done by reproductive organs. As per statistics, patient have nearly 93% high chance to survive if diseases detected during preliminary stage. Conversely, early stage diagnosis hits only 20% due to less accuracy methods. By using Machine Learning identity patterns, detection of ovarian cancer in preliminary stages can be performed in more accurate detected without the guidance of doctors. This paper, introduces the methodology known as Enhanced Max Pooling (EMP) for detecting and classifying an ovarian cancer using advanced machine learning techniques. This methodology discusses the possibilities of using machine learning in the image classification and limitations to overcome.