A Novel Method for identifying tumors in mammography images using with Image mining techniques

Introduction: Breast cancer is the most common cancer and the leading cause of death from cancer among women worldwide. Tumors are in different sizes andshapes so this reason, detection of lesions particularly is very difficult in the early stages of tumor formation. According to official statistics of the National Cancer Institute in America, 10 to 30 percent of breast glands in patients by radiologists in mammography indistinguishable. The aim of this study is to provide a novel method for detecting tumors using mammography with compared in the exist database. Methods: The images have been used in this Study for evaluation is belong to Breast Cancer Surveillance Consortium (BCSC), affiliated to the National Cancer Institute. For analysis of these data we used to Weka software. Output of algorithm in each image had been compared by radiologist detection and this result shown to algorithm have an excellent execute mode. Findings: Comparing the mammogram images with an existence database can be very helpful. This method improved and optimize with neural network. With this method can be very useful and effective in certitude and promptitude. Conclusions:The results of this study showed that using techniques image mining and its integration with the neural network algorithm to accurately identify tumors was 97.4%.

کلیدواژه ها:

Breast cancer ، Image mining ، Data mining ، Neural network ، Weka