First and foremost, I'd like to express my gratitude to the team members who effectively developed this model using various learning algorithms to identify the most proficient one.
The exploration of deep learning methodologies to revolutionize pneumonia diagnosis using chest X-ray images.
Leveraging algorithms such as:
- Convolutional Neural Network (CNN)
- Visual Geometry Group (VGG)
- Residual Networks (ResNet)
- LeNet-5 (LN-5)
Aims to significantly enhance the precision and efficiency of pneumonia detection through robust classification models
Our team is engaged in crafting papers and slide for this research to provide more details about this particular type
The Deep-Chest-Diagnostics Paper
The Deep-Chest-Diagnostics Slide
Feature Extraction
- FeatureExtraction.R
File Model
- CNN.R
- VGG.R
- ResNet.R
- LeNet-5
Evaluated Model
- Evaluate_Model.R
Before proceeding with the code execution, ensure to load the necessary libraries for seamless functionality.
library(keras)
library(tensorflow)
library(Metrics)
library(pROC)
library(tidyverse)
If the required libraries are not already installed, they must be installed first to proceed.
For the results of the evaluated model, our team employs various metrics and techniques to understand the model's predictive capability and generalization ability, such as accuracy, precision, recall, F-score, and the ROC curve
In the README, we solely present ROC curve images example for data split into 40% for training and 60% for testing. For a comprehensive understanding of other results, please refer to the paper for detailed information
Convolutional Neural Network
Visual Geometry Group
Residual Networks
LeNet-5
