Pneumonia Laboratory – Deep Learning Based Pneumonia Prediction System Using Chest X-Rays

Pneumonia is the most common infectious cause of death in children worldwide. In 2017, pneumonia claimed the lives of 808 694 children under the age of five, or 15% of all fatalities in this age group, according to reports from the World Health Organization. Children’s pneumonia can be avoided with simple precautions, treated with economical, low-tech medical treatment, and prevented through vaccination.  Although Sri Lanka devotes a larger portion of its national income to the health sector, there are still some issues that need to be resolved in addition to advocating for free healthcare. In Sri Lanka, the leading cause of pediatric fatalities is pneumonia. Unfortunately, chest x-ray research is subject to human mistake and depends on the reader’s level of knowledge. A big population mass screening is actually a labor-intensive, time-consuming operation that requires significant effort when done manually. The aim of the project is therefore to research and to build a Pneumonia Detector that will help radiologist and doctors to supports and predict pneumonia and also to reduce the number of human errors. Kaggle was used as the system’s key data source for training the machine learning algorithm to construct predictive data models.