Tuberculosis (TB) is one of the most common serious infection in respiratory tract and lungs of human body may lead to death. It is essential to develop an efficient and accurate method to detect TB from chest X-ray images. With the success of deep learning in image classification with subject to health care and treatment, classifying TB in chest X-ray images by deep learning becomes promising. In this project, we decide to experiment on the performance of Deep Neural Network in classifying TB infection presence.
The code resides in Project_TB_detection.ipynb
The code requires the dataset which can be downloaded from the link: LINK
Please download the dataset and keep it in the same directory as the ipynb code file.
This is required as the code has the relative path to the dataset.
The code contains the various experiments to figure out the proper parameters on which the model can be run and produce the output at the end. We achieve the accuracy of greater than 98%