Automatic Mammogram Tumor Detection Using Supervised Learning Method
Chandana Saipriya., Boregowda Dhanalakshmi, S Gnanasoundari., Mercy Therasa. M, J Hemadevi. · International journal of advance research, ideas and innovations in technology · 2017
Breast cancer is the most occupied type of cancer in women that caused the most deaths among women. The early detection of breast cancer is more important for the chances of survival of the patient. This work has mainly four modules: Pre-processing, Segmentation is carried out by Active Contour algorithm and Advanced K-means algorithm, Feature extraction is done by Gray Level Co-occurrence Matrix (GLCM), Expectation Maximization (EM) and Principle Component Analysis (PCA), finally classification is done by Random Forest Classification. To achieve the objective of this work, MIAS (Mammographic Image Analysis Society) and IN breast databases are used as input images. The Accuracy achieved in this system is 95.83%.