A Novel COVID-19 Diagnosis Support System Using the Stacking Approach and Transfer Learning Technique on Chest X-Ray Images

Abstract

COVID-19 is an infectious disease-causing flu-like respiratory problem with various symptoms such as cough or fever, which in severe cases can cause pneumonia. The aim of this paper is to develop a rapid and accurate medical diagnosis support system to detect COVID-19 in chest X-ray images using a stacking approach combining transfer learning techniques and KNN algorithm for selection of the best model. In deep learning, we have multiple approaches for building a classification system for analyzing radiographic images. In this work, we used the transfer learning technique. This approach makes it possible to store and use the knowledge acquired from a pretrained convolutional neural network to solve a new problem. To ensure the robustness of the proposed system for diagnosing patients with COVID-19 using X-ray images, we used a machine learning method called the stacking approach to combine the performances of the many transfer learning-based models. The generated model was trained on a dataset containing four classes, namely, COVID-19, tuberculosis, viral pneumonia, and normal cases. The dataset used was collected from a six-source dataset of X-ray images. To evaluate the performance of the proposed system, we used different common evaluation measures. Our proposed system achieves an extremely good accuracy of 99.23% exceeding many previous related studies.

Lung Disease Classification in CXR Images Using Hybrid Inception-ResNet-v2 Model and Edge Computing

“…While deep networks have the advantage of training models from scratch, they suffer from overfitting on small datasets. Therefore, transfer learning-based models are very popular for COVID-19 detection [ 10 – 12 ]. The development of end-to-end integrated applications based on edge computing, on the other hand, has received relatively little attention.…”

Evaluating multi-state systems reliability with a new improved method

“…In [21], Lin's program that utilizes explicit enumeration to explore f1, f2, f3, f4 is identical to the following code; it employs four loops, and in each loop, flow fi begins at 0 and ends at minimum {M i | ai ∈ pj }, j = 1, 2, 3, ... m as a constraint (3) As it can be noticed, Lin's [21] algorithm enters each loop and stops at the last condition to check if it's false or true. If the condition is satisfied, the algorithm adds the flows (f1, f2, f3, f4) to the set of feasible solutions F, The computation complexity required by running this algorithm using the network of Figure 1 is 6*6*6*6 = 1296 steps (because each flow fi start from 0 to 5).…”

“…Concurrently, there is an urgent need to describe and measure the reliability of a model's prediction based on individual samples. This is particularly true when such models are used in the context of safety-critical areas [1]- [3].…”

Application of Internet of Things on the Healthcare Field Using Convolutional Neural Network Processing

“…Electrocardiogram (ECG) signals are a biophysical indicator of the electrical activity of the heart. It shows how the beating of the heart changes over time [ 9 – 13 ]. Automated systems have a difficult time spotting anomaly.…”