DeePNeu: Robust Detection of Pneumonia Symptoms using Faster R-CNN
Abstract
Every year, more than 150 million people, primarily children under five, develop pneumonia. Various articles present various methods for detecting pneumonia. However, to accurately analyze chest X-ray images, radiologists need expertise field. The traditional techniques remain shortcomings, including the availability of experts, maintenance costs, and expensive tools. Thus, we present a new intelligence method to detect pneumonia images quickly and accurately using the Faster Region Convolutional Neural Network (Faster R-CNN) algorithm. To build our detection model, we collect data, process it first, train it with various parameters to get the best accuracy, and then test it with new data. Based on the experimental results, it was found that this model can accurately detect pneumonia x-ray images marked with bounding boxes. In this model, it is possible to predict the bounding box that is more than what it should be, so NMS is applied to eliminate the prediction of the bounding box that is less precise to increase accuracy
Downloads
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
IJICOM is an open-access journal. Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.