الفهرس | Only 14 pages are availabe for public view |
Abstract Till now mortality among women represent a vast number worldwide. So, studying methods for early detection of cancer tumor with high accuracy and to introduce a new method comfortable for women to analyze, diagnose and treat cancer still represent an importance. The present thesis introduces a method to use breast medical images using CNN to classify breast disease type and location of breast tumors. This thesis presents a combinational technique used for screening breast tumor (e.g., ultrasound (US), MAMO, UWB, Infrared thermography (IRT), and histopathology) and locating tumors in breast with help of AI techniques, also creating CNN application with an interactive environment that can be installed on any operating system for several screening medical tensors that help in classification and localization of breast tumors. Moreover, the proposed method can be used in other bio-signal for classification and to diagnose diseases. An overview of the evolution of deep learning (DL) and a brief idea about the different learning approaches, such as supervised learning, unsupervised learning, to train the neural network, supervised learning utilizes labelled data. To get a better result, hybrid learning combines different architectures pre-trained networks CNN, can be used to implement DL. |