Due to the pandemic brought by the new coronavirus, COVID-19, the search for rapid identification and control of the disease has become a critical issue in health care worldwide. The methods of diagnosis have been developed little by little, and although it is a complement for identification, the use of lungs radiographs has been adopted. In addition, a group of lung diseases can also be identified from radiographs, these are pneumonia and tuberculosis. Image processing is a topic already well studied in the artificial intelligence literature and, more specifically, in the computer vision area. In this work the main object of study is to use convolutional neural network architecture to assist in the classification and possible detection of this group of diseases observable by X-rays. The research was conducted by analyzing three neural models and comparing their results by applying changes in the architecture and deep learning techniques.
-
💾 Create a copy of this repository in a google colab environment.
-
📦 This project uses a kaggle dataset that was downloaded to google drive, if you want to run it you will need to download the dataset to your google drive.
-
🧰 Before running google colab cells, make sure the runtime is configured to run the project using the GPU. ambiente de execução > alterar tipo de ambiante de execução > acelerador de hardware
- Matheus Antonino - MatheusTA
- Matheus Lima - Maath-Lima