Alzheimer Detection via Convnets
- 0 Collaborators
In this project I'm going to develop a software capable of detect Alzheimer disease in magnetic resonance images of different patients and differents disease levels. ...learn more
Project status: Under Development
Intel Technologies
AI DevCloud / Xeon,
Intel Opt ML/DL Framework
Overview / Usage
This project aims to detect Alzheimer's disease in patients with different stages of the disease using convolutional neural networks and magnetic resonance imaging. Our goal is to facilitate the radiologist's task in providing a more specific and accurate diagnosis for the patient. At the moment, our project has purely academic objective, serving as work to conclude my graduation at the University of Brasília, Brazil. The dataset of images is provided by ADNI (Alzheimer's Disease Neuroimaging Initiative) to research.
Methodology / Approach
A metodologia consiste em obter imagens de ressonância magnética de diferentes pacientes e em diferentes estágios da doença de Alzheimer (saudáveis, distúrbios cognitivos leves e doentes), obter as imagens desse exame em seu eixo coronal (uma vez que apresenta de melhor forma as regiões mais afetadas pela doença), e inserí-las em uma rede neural convolucional que seja capaz de ser treinada para diagnosticar novas imagens entre um dos três estágios da doença. As imagens, antes de serem processadas pela rede neural convolucional, são pré processadas, no sentido de melhorar características de contraste/saturação, realce de bordas e remoção de ruídos.
Technologies Used
In this work we use the following tools: Python language (version distributed by Intel DevCloud), Keras Deep Learning Framework (with TensorFlow backend), OpenCV and. SciKit Learn Framework.