Cancer is one of the most immediate deaths causing disease from all other disease. Breast cancer is a leading disease if we see its data globaly. The main goal in cancer prediction is to extract the affected area of ultrasound breast image and predict growth and type in an accurate manner. As in this highly populated world it is needed a system that can use a digital image and scan it in a simple operating manner to extract the affected part of the image. The raw ultrasound images may not be clear to see the affected part due to pixel resolution and noise, so an additional filtering and enhancing the image quality is needed to remove the unwanted features. For extracting the features from the enhanced image, we apply its featured data to genetic algorithms (GA) and store the features into CNN for classification so the process could complete the operation. So, this kind of innovative system is becoming so helpful in case of emergency to treated in a systematic manner. So, it could lead to an improved service speed for populated hospitals and clinics. So, this can help in understanding how GA and CNN are used to extract feature further classified into segments.
Key words: convolution neural network (CNN), genetic algorithm (GA), machine learning
|
