Brain Tumor Detection-based Implementation Using Deep Learning on FPGA

A brain tumor is a mass that grows irregularly in the brain and directly affects human life. This mass appears spontaneously because of the tissue surrounding the brain or skull. Surgical methods are usually preferred for the treatment of brain tumors. Recently, deep learning models in the diagnosis and treatment of diseases in the biomedical field have attracted much interest. In this study, we will choose the pre-trained convolutional neural network model LeNet5 based on deep learning for tumor recognition from brain MRI images. In a first step, the GPU graphics processor will be used to accelerate the learning of convolutional neural networks (CNN). The next step is to design and implement this application, in co-design, on the new PYNQ Z1 demonstration platform based on programmable and reconfigurable FPGA-SoC technologies.

Autorentext
Messaoud, SeifeddineSeifeddine Messaoud holds a doctorate degree in the Electronics and Microelectronics from the Faculty of Sciences of Monastir, University of Monastir, Tunisia. Seifeddine Messaoud conducts research, at the Electronics and Microelectronics Laboratory, on SDN, NFV, Resource Management, 5G, IoT, Industry 4.0 (IIoT), Artificial Intelligence, Deep Learn.