Mamogram görüntülerinde anormalliklerin tespiti için grafik kod tabanlı algoritmaların geliştirilmesi

Nowadays, Mammography is the best available technique and method for early detection of breast cancer. The most common abnormalities in breast that may indicate breast cancer are appeared masses. Also, there are some signs that can lead to breast cancer diagnosis, such as architectural distortion and bilateral asymmetry. Generally, an algorithm is used to detect and classify breast cancer in mammography images. Three stages are presented: (1) Image preprocessing and mass segmentation, (2) feature selection and extraction and (3) classification. In the first stage which is preprocessing stage some image processing techniques can be used to enhance the image then segment the suspected mass inside the mammography image of the breast. Image pruning, applying smoothing and gaussian filters, thresholding and morphological operations are exaples and steps for image processing stage. The second stage is for selecting and extracting some features from segmented mass to use these features in the next and last step which is classification. Generally the features are related to the Intensity of the mass compared to the other parts of breast and background of the image. The last stage is classification and as it is clear from its name that it is a system for classifying the mass depending on the extracted features of the segmented mass from the previous step. There are some different algorithms for classifying such as support vector machine (SVM) and artificial neural network (ANN). In this system the ANN machine learning tool is used for the clasification stage to get the best result.