A Deep Analysis Study on Leverage Convolutional Neural Networks Method for Pneumonia Detection System using Machine Learning
Shrestha Majumder, Dr R.NaveenkumarThe use of data mining and machine learning has become essential for the detection and prevention of various diseases. For children under five, interstitial lung diseases like pneumonia are the main cause of death. Children under five every year are affected by each day for various reasons. This includes around the maximum number ofnewborn babies. Almost all the maximum deaths are preventable. According to a UNICEF report, there are more than 1,400 instances of pneumonia per 100,000 children worldwide or one case for every 71 children annually. The majority of the affected kids were less than two. The healing process for children with pneumonia can be accelerated with prompt diagnosis. To effectively identify pneumonic lungs from chest X-rays, we have applied convolutional neural network models in this study for a better result. Medical professionals can use these models to treat pneumonia in the actual world. The first, second, third and fourth model consists of two convolutional layers. The first model achieves an accuracy of 89.74%, the second one reaches an accuracy of 85.26%. Furthermore, recall and F1 scores are calculated from the confusion matrix of each model for better evaluation.